├── CCCascadeClassifier.cpp
├── CCCascadeClassifier.h
├── FDImage.cpp
├── FDImage.h
├── NeuralNetWork.cpp
├── README.md
├── cnn.cpp
├── cnn.h
├── data
    ├── .DS_Store
    ├── cascadeClassifier.txt
    ├── outColor1.mp4
    ├── outColor2.mp4
    ├── resultMatrix.txt
    ├── zInput12
    │   └── theanocnn.json
    ├── zInput20
    │   └── theanocnn.json
    ├── zInput24
    │   └── theanocnn.json
    ├── zInput40
    │   ├── .DS_Store
    │   ├── theanocnn.json
    │   ├── 参数1
    │   │   ├── .DS_Store
    │   │   └── theanocnn.json
    │   ├── 参数2
    │   │   ├── .DS_Store
    │   │   └── theanocnn.json
    │   └── 参数3
    │   │   ├── .DS_Store
    │   │   └── theanocnn.json
    └── zKernel20-50
    │   └── theanocnn.json
├── featuremap.cpp
├── featuremap.h
├── head
    ├── .DS_Store
    ├── libopencv_calib3d.2.4.13.dylib
    ├── libopencv_calib3d.2.4.dylib
    ├── libopencv_calib3d.dylib
    ├── libopencv_contrib.2.4.13.dylib
    ├── libopencv_contrib.2.4.dylib
    ├── libopencv_contrib.dylib
    ├── libopencv_core.2.4.13.dylib
    ├── libopencv_core.2.4.dylib
    ├── libopencv_core.dylib
    ├── libopencv_features2d.2.4.13.dylib
    ├── libopencv_features2d.2.4.dylib
    ├── libopencv_features2d.dylib
    ├── libopencv_flann.2.4.13.dylib
    ├── libopencv_flann.2.4.dylib
    ├── libopencv_flann.dylib
    ├── libopencv_gpu.2.4.13.dylib
    ├── libopencv_gpu.2.4.dylib
    ├── libopencv_gpu.dylib
    ├── libopencv_highgui.2.4.13.dylib
    ├── libopencv_highgui.2.4.dylib
    ├── libopencv_highgui.dylib
    ├── libopencv_imgproc.2.4.13.dylib
    ├── libopencv_imgproc.2.4.dylib
    ├── libopencv_imgproc.dylib
    ├── libopencv_legacy.2.4.13.dylib
    ├── libopencv_legacy.2.4.dylib
    ├── libopencv_legacy.dylib
    ├── libopencv_ml.2.4.13.dylib
    ├── libopencv_ml.2.4.dylib
    ├── libopencv_ml.dylib
    ├── libopencv_nonfree.2.4.13.dylib
    ├── libopencv_nonfree.2.4.dylib
    ├── libopencv_nonfree.dylib
    ├── libopencv_objdetect.2.4.13.dylib
    ├── libopencv_objdetect.2.4.dylib
    ├── libopencv_objdetect.dylib
    ├── libopencv_ocl.2.4.13.dylib
    ├── libopencv_ocl.2.4.dylib
    ├── libopencv_ocl.dylib
    ├── libopencv_photo.2.4.13.dylib
    ├── libopencv_photo.2.4.dylib
    ├── libopencv_photo.dylib
    ├── libopencv_stitching.2.4.13.dylib
    ├── libopencv_stitching.2.4.dylib
    ├── libopencv_stitching.dylib
    ├── libopencv_superres.2.4.13.dylib
    ├── libopencv_superres.2.4.dylib
    ├── libopencv_superres.dylib
    ├── libopencv_ts.a
    ├── libopencv_video.2.4.13.dylib
    ├── libopencv_video.2.4.dylib
    ├── libopencv_video.dylib
    ├── libopencv_videostab.2.4.13.dylib
    ├── libopencv_videostab.2.4.dylib
    ├── libopencv_videostab.dylib
    ├── pkgconfig
    │   └── opencv.pc
    └── python2.7
    │   └── site-packages
    │       ├── cv.py
    │       └── cv2.so
├── hiddenLayer.cpp
├── hiddenLayer.h
├── logisticRegression.cpp
├── logisticRegression.h
├── main.cpp
├── mlp.cpp
├── mlp.h
├── neuralNetwork.h
├── neuralbase.cpp
├── neuralbase.h
├── poollayer.cpp
├── poollayer.h
├── stdafx.cpp
├── stdafx.h
├── targetver.h
├── util.cpp
└── util.h


/CCCascadeClassifier.cpp:
--------------------------------------------------------------------------------
  1 | #include "stdafx.h"
  2 | #include <fstream>
  3 | #include <vector>
  4 | #include <math.h>
  5 | #include <string>
  6 | #include <algorithm>
  7 | #include <set>
  8 | using namespace std;
  9 | 
 10 | #include "FDImage.h"
 11 | #include "CCCascadeClassifier.h"
 12 | 
 13 | #include "mlp.h"  
 14 | #include "util.h"  
 15 | // #include "testinherit.h"  
 16 | #include "neuralNetwork.h" 
 17 | #include <opencv2/core/core.hpp>
 18 | #include <opencv2/highgui/highgui.hpp>
 19 | #include <opencv2/opencv.hpp>
 20 | // 新版本写在下面文件中：
 21 | #include <opencv2/nonfree/features2d.hpp>
 22 | //#include "opencv2/features2d/features2d.hpp"
 23 | #include<opencv2/legacy/legacy.hpp>
 24 | 
 25 | using namespace cv;
 26 | 
 27 | #ifdef _DEBUG
 28 | #define new DEBUG_NEW
 29 | #undef THIS_FILE
 30 | static char THIS_FILE[] = __FILE__;
 31 | #endif
 32 | 
 33 | CCCascadeClassifier::CCCascadeClassifier():count(0),indexBoosted(0)
 34 | {
 35 |     // resultsGroup.resize(9);
 36 | }
 37 | 
 38 | CCCascadeClassifier::~CCCascadeClassifier()
 39 | {
 40 |     Clear();
 41 | }
 42 | 
 43 | void CCCascadeClassifier::Clear()
 44 | {
 45 |     count = 0;
 46 | }
 47 | 
 48 | CCCascadeClassifier& CCCascadeClassifier::operator=(const CCCascadeClassifier& source)
 49 | {
 50 |     Clear();
 51 |     count = source.count;
 52 |     return *this;
 53 | }
 54 | 
 55 | void CCCascadeClassifier::DrawResults(FDImage& image, vector<Rect>& results) const
 56 | {
 57 | 	int i;
 58 | 	unsigned int k;
 59 | 	int x1,x2,y1,y2;
 60 | 
 61 | 	for(k=0;k<results.size();k++)
 62 | 	{
 63 | 		y1 = (results[k].y>=0)?results[k].y:0;
 64 | 		y1 = (results[k].y<image.height)?results[k].y:(image.height-1);
 65 | 		y2 = (results[k].y+results[k].height>=0)?results[k].y+results[k].height:0;
 66 | 		y2 = (results[k].y+results[k].height<image.height)?results[k].y+results[k].height:(image.height-1);
 67 | 		x1 = (results[k].x>=0)?results[k].x:0;
 68 | 		x1 = (results[k].x<image.width)?results[k].x:(image.width-1);
 69 | 		x2 = (results[k].x+results[k].width>=0)?results[k].x+results[k].width:0;
 70 | 		x2 = (results[k].x+results[k].width<image.width)?results[k].x+results[k].width:(image.width-1);
 71 | 		for(i=y1;i<=y2;i++) 
 72 | 		{
 73 | 			image.data[i][x1] = 255;
 74 | 			image.data[i][x2] = 255;
 75 | 		}
 76 | 		for(i=x1;i<=x2;i++)
 77 | 		{
 78 | 			image.data[y1][i] = 255;
 79 | 			image.data[y2][i] = 255;
 80 | 		}
 81 | 		results[k].x = x1;
 82 | 		results[k].y = y1;
 83 | 		results[k].width = x2 - results[k].x;
 84 | 		results[k].height = y2 - results[k].y;
 85 | 	}
 86 | }
 87 | 
 88 | void CCCascadeClassifier::DrawResults2(FDImage& image,const vector<rectScore>& results) const
 89 | {
 90 | 	int i;
 91 | 	unsigned int k;
 92 | 	int x1,x2,y1,y2;
 93 | 
 94 | 	for(k=0;k<results.size();k++)
 95 | 	{
 96 | 		y1 = (results[k].rect.y>=0)?results[k].rect.y:0;
 97 | 		y1 = (results[k].rect.y<image.height)?results[k].rect.y:(image.height-1);
 98 | 		y2 = (results[k].rect.y+results[k].rect.height>=0)?results[k].rect.y+results[k].rect.height:0;
 99 | 		y2 = (results[k].rect.y+results[k].rect.height<image.height)?results[k].rect.y+results[k].rect.height:(image.height-1);
100 | 		x1 = (results[k].rect.x>=0)?results[k].rect.x:0;
101 | 		x1 = (results[k].rect.x<image.width)?results[k].rect.x:(image.width-1);
102 | 		x2 = (results[k].rect.x+results[k].rect.width>=0)?results[k].rect.x+results[k].rect.width:0;
103 | 		x2 = (results[k].rect.x+results[k].rect.width<image.width)?results[k].rect.x+results[k].rect.width:(image.width-1);
104 | 		// ’‚µÿ∑ΩæÕ «ª≠≥ˆ¿¥»À¡≥øÚ£¨∞—»À¡≥µƒŒª÷√¥Ú…œ∞◊µ„
105 | 		for(i=y1;i<=y2;i++) 
106 | 		{
107 | 			image.data[i][x1] = 255;
108 | 			image.data[i][x2] = 255;
109 | 		}
110 | 		for(i=x1;i<=x2;i++)
111 | 		{
112 | 			image.data[y1][i] = 255;
113 | 			image.data[y2][i] = 255;
114 | 		}
115 | 	}
116 | }
117 | 
118 | 
119 | // ¥À¥¶µƒx,y±Ì æµƒ «x––y¡–µƒ÷µ,∂¯≤ª «◊¯±Í£¨◊™ªØŒ™openCV÷–µƒ◊¯±ÍŒ™£®y£¨x£©
120 | void CCCascadeClassifier::getInputData(FDImage &im,double* data,int x,int y,int rh,int rw, REAL wRatio)
121 | {
122 | 	if (wRatio == 1)
123 | 	{
124 | 		for(int i=0; i<rh; i++)
125 | 		{
126 | 			for(int j=0; j<rw; j++)
127 | 			{
128 | 				data[i*rw+j] = im.data[i+x][j+y]/256.0;
129 | 			}
130 | 		}
131 | 	}
132 | 	else
133 | 	{
134 | 		FDImage image;
135 | 		Rect rect;
136 | 		// openCV÷–µƒ◊¯±Í◊™ªØ°£
137 | 		rect.x = y;
138 | 		rect.y = x;
139 | 		rect.height = rh*wRatio;
140 | 		rect.width = rw*wRatio;
141 | 		im.CutToLxL(image,rect,40);
142 | 		for(int i=0; i<rh/wRatio; i++)
143 | 		{
144 | 			for(int j=0; j<rw/wRatio; j++)
145 | 			{
146 | 				data[i*rw+j] = image.data[i][j]/256.0;
147 | 			}
148 | 		}
149 | 	}
150 | }
151 | 
152 | vector<Rect> CCCascadeClassifier::FaceDetectWithRet_ScaleIamge(FDImage& original,const string filename)
153 | {
154 | 	int sx = 40;
155 | 	int sy = 40;
156 | 
157 | 	ratio = 1.0;
158 | 	original.Resize(procface,ratio);
159 | 	results.clear();
160 |     
161 |     // 步长
162 |     // int stripSize = sx/5;
163 |     int stripSize = 8;
164 |     
165 | 	while((procface.height+1>sx+1) && (procface.width+1>sy+1))
166 | 	{
167 | 		for(int i=0,size_x=procface.height+1-sx;i<size_x;i+=stripSize)
168 | 			for(int j=0,size_y=procface.width+1-sy;j<size_y;j+=stripSize){
169 | 				
170 | 				getInputData(procface,data40,i,j,sx,sy,1);
171 | 
172 | 				// 40*40 CNNºÏ≤‚¥∞ø⁄
173 | 				int iResult40 = TestCnnTheano(sx, 2, data40, T_time);
174 | 
175 | 				if (iResult40 == 1)
176 | 				{
177 |                     /*
178 | 				    int iWidth24 = 24;
179 | 					// openCV÷–µƒ◊¯±Í◊™ªØ°£
180 | 					rect40.x = j;
181 | 					rect40.y = i;
182 | 					rect40.height = sx;
183 | 					rect40.width = sy;
184 | 					procface.CutToLxL(image,rect40,iWidth24);
185 | 					for(int m=0; m<iWidth24; m++)
186 | 					{
187 | 						for(int n=0; n<iWidth24; n++)
188 | 						{
189 | 							data24[m*iWidth24+n] = image.data[m][n]/256.0;
190 | 						}
191 | 					}
192 | 					// 24*24 CNNºÏ≤‚¥∞ø⁄
193 | 					int iResult24 = TestCnnTheano24x24(iWidth24, 2, data24, T_time);
194 | 
195 | 					if (iResult24 == 1)
196 | 					{*/
197 |                         /*
198 | 						//FDImage image;
199 | 						//Rect rect40;
200 | 						int iWidth20 = 20;
201 | 						// openCV÷–µƒ◊¯±Í◊™ªØ°£
202 | 						rect40.x = j;
203 | 						rect40.y = i;
204 | 						rect40.height = sx;
205 | 						rect40.width = sy;
206 | 						procface.CutToLxL(image,rect40,iWidth20);
207 | 						for(int i=0; i<iWidth20; i++)
208 | 						{
209 | 							for(int j=0; j<iWidth20; j++)
210 | 							{
211 | 								data20[i*iWidth20+j] = image.data[i][j]/256.0;
212 | 							}
213 | 						}
214 | 						// 20*20 CNNºÏ≤‚¥∞ø⁄
215 | 						int iResult20 = TestCnnTheano20x20(iWidth20, 2, data20, T_time);
216 | 
217 | 						if (iResult20 == 1)
218 |                          */
219 | 
220 | 						{
221 |                             
222 | 							const REAL r = 1.0/ratio;
223 | 							rect.x = (long)(j*r);rect.y = (long)(i*r);
224 | 							rect.width = (long)((j+sx)*r)-rect.x;rect.height = (long)((i+sy)*r)-rect.y;
225 | 							results.push_back(rect);
226 | 						}
227 | 					//}
228 | 				}
229 | 			}
230 |             // 0.8, 0.6
231 | 			ratio = ratio*0.8;
232 | 			procface.Clear();
233 | 			original.Resize(procface,ratio);
234 | 	}
235 | 
236 |     if(results.size()){
237 | 	    // PostProcess(results,100);
238 | 	    // PostProcess(results,4);
239 |         
240 |         PostProcess(results,5);
241 | 	    //PostProcess(results,0);
242 |         
243 | 	    // DrawResults(original,results);
244 | 	    // original.Save(filename+"_result.JPG");
245 |         
246 |         //nms(results,res,0.6f);
247 |         //results = res;
248 |         
249 | 	    
250 |     }
251 |     return results;
252 | }
253 | 
254 | inline int SizeOfRect(const Rect& rect)
255 | {
256 | 	return rect.height*rect.width;
257 | }
258 | // PostProcess的参数：所有检测窗口 和 combine_min值。
259 | void CCCascadeClassifier::PostProcess(vector<Rect>& result,const int combine_min)
260 | {
261 | 	vector<Rect> res1;
262 | 	vector<Rect> resmax;
263 | 	vector<int> res2;
264 | 	bool yet;
265 | 	Rect rectInter,rectUnion;
266 | 
267 | 	for(unsigned int i=0,size_i=result.size();i<size_i;i++){
268 | 		yet = false;
269 | 		Rect& result_i = result[i];
270 | 		for(unsigned int j=0,size_r=res1.size();j<size_r;j++){
271 | 			Rect& resmax_j = resmax[j];
272 |             rectInter = result_i & resmax_j;
273 |             rectUnion = result_i | resmax_j;
274 | 			if( rectInter.area() != 0 && rectUnion.area() != result_i.area()){
275 | 			//	if((double)SizeOfRect(rectInter)/SizeOfRect(rectUnion) > 0.6){
276 | 				if((double)SizeOfRect(rectInter)/SizeOfRect(result_i) > 0.6 &&
277 | 					(double)SizeOfRect(rectInter)/SizeOfRect(resmax_j) > 0.6 ){
278 | 					Rect& res1_j = res1[j];
279 | 					resmax_j= resmax_j | result_i;
280 |                     //int temp = res1_j.y;
281 |                     res1_j.y += result_i.y;
282 |                     //res1_j.height = res1_j.height + temp + result_i.height + result_i.y - res1_j.y;
283 |                     res1_j.height += result_i.height;
284 |                     //int tmp = res1_j.x;
285 | 					res1_j.x += result_i.x;
286 |                     //res1_j.width = res1_j.width + tmp + result_i.width + result_i.x - res1_j.x;
287 | 					res1_j.width += result_i.width;
288 | 					res2[j]++;
289 | 					yet = true;
290 | 					break;
291 | 				}
292 | 			}
293 | 		}
294 | 		if(yet==false){
295 | 			res1.push_back(result_i);
296 | 			resmax.push_back(result_i);
297 | 			res2.push_back(1);
298 | 		}
299 | 	}
300 | 
301 | 	for(unsigned int i=0,size=res1.size();i<size;i++){
302 | 		const int count = res2[i];
303 | 		Rect& res1_i = res1[i];
304 |         //int tmp1 = res1_i.y;
305 | 		res1_i.y /= count;
306 |         //res1_i.height = (res1_i.height + tmp1)/count - res1_i.y;
307 | 		res1_i.height /= count;
308 |         //int tmp2 = res1_i.x;
309 | 		res1_i.x /= count;
310 |         //res1_i.width = (res1_i.width + tmp2)/count - res1_i.x;
311 | 		res1_i.width /= count;
312 | 	}
313 | 
314 | 	vector<Rect> result1,result2;vector<int> res3,res4;
315 | 	for(int i=0,size=res1.size();i<size;i++){
316 | 		if(res2[i]>combine_min){
317 | 			result2.push_back(res1[i]);
318 | 			res4.push_back(res2[i]);
319 | 		}
320 | 	}
321 |     
322 | 	bool bcan;
323 | 	while(true){
324 | 		bcan = false;
325 | 		vector<int> toDel;
326 | 		result1.clear();result1 = result2;
327 | 		res3.clear();res3 = res4;
328 | 		for(int i=0,size_i=result1.size();i<size_i;i++)
329 | 			toDel.push_back(1);
330 | 
331 | 		for(int i=0,size_i=result1.size();i<size_i;i++)
332 | 		{
333 | 			Rect& result_i = result1[i];
334 | 			for(unsigned int j=i+1,size_j=result1.size();j<size_j;j++)
335 | 			{
336 | 				Rect& result_j = result1[j];
337 |                 rectInter = result_i & result_j;
338 | 				if(rectInter.area() != 0)
339 | 				{
340 | 					double areaInter = (double)SizeOfRect(rectInter),areaI = (double)SizeOfRect(result_i),areaJ=(double)SizeOfRect(result_j);
341 | 					//double s1 = (double)(areaInter/areaI)/(areaInter/areaJ);
342 | 					//double s2 = (double)(areaInter/areaJ)/(areaInter/areaI);
343 | 					double s1 = areaInter/areaI;
344 | 					double s2 = areaInter/areaJ;
345 | 					if( s1/s2 >  1 || s2/s1 > 1 || areaInter/areaI > 0.6 || areaInter/areaJ > 0.6){
346 | 						if(res3[i] > res3[j])
347 | 							toDel[j] = 0;
348 | 						else
349 | 							toDel[i] = 0;
350 | 						bcan = true;
351 | 					}
352 | 				}
353 | 			}
354 | 		}
355 | 		if(bcan == false)
356 | 			break;
357 | 		result2.clear();res4.clear();
358 | 		for(unsigned int i=0,size=result1.size();i<size;i++) 
359 | 			if(toDel[i] == 1){
360 | 				result2.push_back(result1[i]);
361 | 				res4.push_back(res3[i]);
362 | 			}
363 | 	}
364 |     
365 | 	result.clear();
366 | 	result = result2;
367 | }
368 | 
369 | void CCCascadeClassifier::nms(
370 |          const std::vector<cv::Rect>& srcRects,
371 |          std::vector<cv::Rect>& resRects,
372 |          float thresh
373 |          )
374 | {
375 |     resRects.clear();
376 |     
377 |     const size_t size = srcRects.size();
378 |     if (!size)
379 |     {
380 |         return;
381 |     }
382 |     
383 |     // Sort the bounding boxes by the bottom - right y - coordinate of the bounding box
384 |     std::multimap<int, size_t> idxs;
385 |     for (size_t i = 0; i < size; ++i)
386 |     {
387 |         idxs.insert(std::pair<int, size_t>(srcRects[i].br().y, i));
388 |     }
389 |     
390 |     // keep looping while some indexes still remain in the indexes list
391 |     while (idxs.size() > 0)
392 |     {
393 |         // grab the last rectangle
394 |         auto lastElem = --std::end(idxs);
395 |         const cv::Rect& rect1 = srcRects[lastElem->second];
396 |         
397 |         resRects.push_back(rect1);
398 |         
399 |         idxs.erase(lastElem);
400 |         
401 |         for (auto pos = std::begin(idxs); pos != std::end(idxs); )
402 |         {
403 |             // grab the current rectangle
404 |             const cv::Rect& rect2 = srcRects[pos->second];
405 |             
406 |             float intArea = (rect1 & rect2).area();
407 |             float unionArea = rect1.area() + rect2.area() - intArea;
408 |             float overlap = intArea / unionArea;
409 |             
410 |             // if there is sufficient overlap, suppress the current bounding box
411 |             if (overlap > thresh)
412 |             {
413 |                 pos = idxs.erase(pos);
414 |             }
415 |             else
416 |             {
417 |                 ++pos;
418 |             }
419 |         }
420 |     }
421 | }
422 | 


--------------------------------------------------------------------------------
/CCCascadeClassifier.h:
--------------------------------------------------------------------------------
 1 | 
 2 | #include <opencv2/core/core.hpp>
 3 | #include <opencv2/highgui/highgui.hpp>
 4 | #include <opencv2/opencv.hpp>
 5 | // 新版本写在下面文件中：
 6 | #include <opencv2/nonfree/features2d.hpp>
 7 | //#include "opencv2/features2d/features2d.hpp"
 8 | #include<opencv2/legacy/legacy.hpp>
 9 | 
10 | #include <vector>
11 | using namespace std;
12 | using namespace cv;
13 | 
14 | class FDImage;
15 | 
16 | struct rectScore{
17 | 	Rect rect;
18 | 	float score;
19 | };
20 | 
21 | 
22 | struct CCCascadeClassifier
23 | {
24 | 	int count;
25 | 	int indexBoosted;
26 | 	CCCascadeClassifier();
27 | 	~CCCascadeClassifier();
28 | 	void Clear(void);
29 | 	CCCascadeClassifier& operator=(const CCCascadeClassifier& source);
30 | 
31 | 	vector<Rect>  FaceDetectWithRet_ScaleIamge(FDImage& original,const string filename);
32 | 	
33 |     void DrawResults(FDImage& image, vector<Rect>& results) const;
34 | 	void DrawResults2(FDImage& image, const vector<rectScore>& results) const;
35 |     void PostProcess(vector<Rect>& result, const int combine_min);
36 |     void nms(
37 |              const std::vector<cv::Rect>& srcRects,
38 |              std::vector<cv::Rect>& resRects,
39 |              float thresh
40 |              );
41 | 
42 | 
43 | 
44 | 	double Expression_Recognition(FDImage& original,const string filename);
45 | 
46 | 	void getInputData(FDImage &im,double* data,int x,int y,int rh,int rw, REAL ratio);
47 | 
48 | 	FDImage procface;
49 | 	double value;
50 | 	double data40[1600];
51 | 	double data24[576];
52 | 	double data20[400];
53 | 	double T_time;
54 | 	Rect rect;
55 | 	REAL ratio;
56 | 	vector<Rect> results;
57 |     vector<Rect> res;
58 | 	FDImage image;
59 | 	Rect rect40;
60 | };
61 | 
62 | 


--------------------------------------------------------------------------------
/FDImage.cpp:
--------------------------------------------------------------------------------
  1 | 
  2 | #pragma   comment(lib,   "vfw32.lib ")
  3 | #pragma comment (lib , "comctl32.lib")
  4 | 
  5 | #include <opencv2/core/core.hpp>
  6 | #include <opencv2/highgui/highgui.hpp>
  7 | #include <opencv2/opencv.hpp>
  8 | // 新版本写在下面文件中：
  9 | #include <opencv2/nonfree/features2d.hpp>
 10 | //#include "opencv2/features2d/features2d.hpp"
 11 | #include<opencv2/legacy/legacy.hpp>
 12 | #include <algorithm>
 13 | #include <fstream>
 14 | using namespace std;
 15 | 
 16 | #include "FDImage.h"
 17 | #include "CCCascadeClassifier.h"
 18 | 
 19 | #ifdef _DEBUG
 20 | #define new DEBUG_NEW
 21 | #undef THIS_FILE
 22 | static char THIS_FILE[] = __FILE__;
 23 | #endif
 24 | 
 25 | FDImage::FDImage():height(0),width(0),data(NULL),label(-1),variance(0.0)
 26 | {
 27 | 	for(int i=0;i<height;i++){
 28 | 			data[i] = new REAL;
 29 | 		}
 30 | 		buf = new REAL;
 31 | 
 32 | }
 33 | 
 34 | FDImage::~FDImage()
 35 | {
 36 | 	delete [] img;
 37 | 	Clear();
 38 | }
 39 | 
 40 | void FDImage::Clear(void)
 41 | {
 42 | 
 43 | 	if(data == NULL)
 44 | 		assert(buf == NULL);
 45 | 	else
 46 | 	{
 47 | 		assert(buf != NULL);
 48 | 		for(int i=0;i<height;i++){
 49 | 			data[i] = NULL;
 50 | 		}
 51 | 		delete[] data;	data = NULL;
 52 | 		delete[] buf;  	buf = NULL;
 53 | 	
 54 | 		height = width = 0;
 55 | 		variance = 0.0;
 56 | 		label = -1;
 57 | 	}	
 58 | 
 59 | }
 60 | 
 61 | void FDImage::cleartmp()
 62 | {
 63 | 	if(data == NULL)
 64 | 		assert(buf == NULL);
 65 | 	else
 66 | 	{
 67 | 		assert(buf != NULL);
 68 | 		for(int i=0;i<height;i++){
 69 | 			data[i] = NULL;
 70 | 		}
 71 | 		delete[] data;	data = NULL;
 72 | 		delete[] buf;  	buf = NULL;
 73 | 	}
 74 | 	//if(data  != NULL){
 75 | 	//	for(int i=0;i<height;i++)	data[i] = NULL;
 76 | 	//	delete[] data;	data = NULL;
 77 | 	//	delete[] buf;	buf = NULL;
 78 | 	//}
 79 | 	//if(m_gradientImage != NULL){
 80 | 	//	delete [] m_gradientImage;
 81 | 	//	m_gradientImage=NULL;
 82 | 	//}
 83 | }
 84 | void FDImage::Copy(const FDImage& source)
 85 | 	// the ONLY way to make a copy of 'source' to this image
 86 | {
 87 | 	assert(source.height > 0);
 88 | 	assert(source.width > 0);
 89 | 	if(&source == this)	return;
 90 | 	SetSize(cvSize(source.height,source.width));
 91 | 	label = source.label;
 92 | 	memcpy(buf,source.buf,sizeof(REAL)*height*width);
 93 | }
 94 | 
 95 | void FDImage::Load(IplImage* temp)
 96 | {
 97 | 	img = temp;
 98 | 	SetSize(cvSize(img->height,img->width));
 99 | 	for(int i=0,ih=img->height,iw=img->width;i<ih;i++)
100 | 	{
101 | 		REAL* pdata = data[i];
102 | 		unsigned char* pimg = reinterpret_cast<unsigned char*>(img->imageData+img->widthStep*i);
103 | 		for(int j=0;j<iw;j++) pdata[j] = pimg[j];
104 | 	}
105 | }
106 | 
107 | void FDImage::SetSize(const CvSize size)
108 | 	// 'size' is the new size of the image, if necessary, memory is reallocated
109 | 	// size.cx is the new height and size.cy is the new width
110 | {
111 | 	if((size.height == height) && (size.width == width) && (buf != NULL) &&(data != NULL) ) return; 
112 | 	assert(size.height >= 0); assert(size.width >= 0);
113 | 	//Clear();
114 | 	height = size.height;	width = size.width;
115 | 	buf = new REAL[height*width]; assert(buf != NULL);
116 | 	data = new REAL*[height];	assert(data != NULL);
117 | 	for(int i=0;i<height;i++)	data[i] = &buf[i*width];
118 | 
119 | }
120 | 
121 | FDImage& FDImage::operator=(const FDImage& source)
122 | {
123 | 	SetSize(cvSize(source.height,source.width));
124 | 	//memcpy(data,source.data,sizeof(double)*height*width);
125 | 	label = source.label;
126 | 
127 | 	return *this;
128 | }
129 | 
130 | void FDImage::Resize(FDImage &result, REAL ratio) const
131 | {
132 | 	result.SetSize(cvSize(int(height*ratio),int(width*ratio)));
133 | 	ratio = 1/ratio;
134 | 	for(int i=0,rh=result.height,rw=result.width;i<rh;i++)
135 | 		for(int j=0;j<rw;j++) {
136 | 			int x0,y0;
137 | 			REAL x,y,fx0,fx1;
138 | 			x = j*ratio; y = i*ratio;
139 | 			x0 = (int)(x);
140 | 			y0 = (int)(y);
141 | 
142 | 			//by Jianxin Wu  
143 | 			//1. The conversion of float to int in C is towards to 0 point, i.e. the floor function for positive numbers, and ceiling function for negative numbers.
144 | 			//2. We only make use of ratio<1 in this applicaiton, and all numbers involved are positive.
145 | 			//Using these, we have 0<=x<=height-1 and 0<=y<=width-1. Thus, boundary conditions check is not necessary.
146 | 			//In languages other than C/C++ or ratio>=1, take care. 
147 | 			if (x0 == width-1) x0--;
148 | 			if (y0 == height-1) y0--;
149 | 
150 | 			x = x - x0; y = y - y0;
151 | 
152 | 			fx0 = data[y0][x0] + x*(data[y0][x0+1]-data[y0][x0]);
153 | 			fx1 = data[y0+1][x0] + x*(data[y0+1][x0+1]-data[y0+1][x0]);
154 | 
155 | 			result.data[i][j] = fx0 + y*(fx1-fx0);
156 | 		}
157 | }
158 | 
159 | void FDImage::CutToLxL(FDImage &result, CvRect rect, int iL) const
160 | {
161 | 	CvSize dst_cvsize; IplImage* pdst = 0;
162 | 	dst_cvsize.width = iL; //ƒø±ÍÕºœÒøÌ∂»  
163 | 	dst_cvsize.height = iL; //ƒø±ÍÕºœÒ∏fl∂» 
164 | 	pdst = cvCreateImage(dst_cvsize, IPL_DEPTH_8U, 1);
165 | 	
166 | 	// Ω´ƒ⁄¥Ê÷–µƒÕº∆¨ –¥µΩ opencv Ω·ππµƒÕº∆¨÷–
167 | 	IplImage* tsimg;
168 | 	tsimg = cvCreateImage(cvSize(width,height),IPL_DEPTH_8U,1);
169 | 	for(int i=0,ih=tsimg->height,iw=tsimg->width;i<ih;i++)
170 | 	{
171 | 		REAL* pdata = data[i];
172 | 		unsigned char* pimg = reinterpret_cast<unsigned char*>(tsimg->imageData+tsimg->widthStep*i);
173 | 		for(int j=0;j<iw;j++) pimg[j] = (unsigned char)pdata[j];
174 | 	}
175 | 	// ¿˚”√ opencv ∫Ø ˝Ω¯––Àı∑≈
176 | 	cv::Mat img(tsimg);
177 | 	cv::Mat temp;
178 | 	img(rect).copyTo(temp);
179 | 	IplImage *src = cvCreateImage(cvSize(temp.cols,temp.rows),8,1);
180 | 	
181 | 	cvResize(src,pdst,CV_INTER_LINEAR);
182 | 	// ‘⁄øΩµΩƒø±Í∂‘œÛ÷–
183 | 
184 | 	result.SetSize(cvSize(pdst->height,pdst->width));
185 | 	for(int i=0,ih=pdst->height,iw=pdst->width;i<ih;i++)
186 | 	{
187 | 		REAL* pdata = result.data[i];
188 | 		unsigned char* pimg = reinterpret_cast<unsigned char*>(pdst->imageData+pdst->widthStep*i);
189 | 		for(int j=0;j<iw;j++) pdata[j] = pimg[j];
190 | 	}
191 | 
192 | 	cvReleaseImage(&pdst);cvReleaseImage(&tsimg);
193 | 
194 | }
195 | 
196 | void FDImage::Resize240x40(FDImage &result) const
197 | {
198 | 	CvSize dst_cvsize; IplImage* pdst = 0;
199 | 	dst_cvsize.width = 40; //ƒø±ÍÕºœÒøÌ∂»  
200 | 	dst_cvsize.height = 40; //ƒø±ÍÕºœÒ∏fl∂» 
201 | 	pdst = cvCreateImage(dst_cvsize, IPL_DEPTH_8U, 1);
202 | 	
203 | 	// Ω´ƒ⁄¥Ê÷–µƒÕº∆¨ –¥µΩ opencv Ω·ππµƒÕº∆¨÷–
204 | 	IplImage* tsimg;
205 | 	tsimg = cvCreateImage(cvSize(width,height),IPL_DEPTH_8U,1);
206 | 	for(int i=0,ih=tsimg->height,iw=tsimg->width;i<ih;i++)
207 | 	{
208 | 		REAL* pdata = data[i];
209 | 		unsigned char* pimg = reinterpret_cast<unsigned char*>(tsimg->imageData+tsimg->widthStep*i);
210 | 		for(int j=0;j<iw;j++) pimg[j] = (unsigned char)pdata[j];
211 | 	}
212 | 	// ¿˚”√ opencv ∫Ø ˝Ω¯––Àı∑≈
213 | 	cvResize(tsimg,pdst,CV_INTER_LINEAR);
214 | 	// ‘⁄øΩµΩƒø±Í∂‘œÛ÷–
215 | 
216 | 	result.SetSize(cvSize(pdst->height,pdst->width));
217 | 	for(int i=0,ih=pdst->height,iw=pdst->width;i<ih;i++)
218 | 	{
219 | 		REAL* pdata = result.data[i];
220 | 		unsigned char* pimg = reinterpret_cast<unsigned char*>(pdst->imageData+pdst->widthStep*i);
221 | 		for(int j=0;j<iw;j++) pdata[j] = pimg[j];
222 | 	}
223 | 
224 | 	cvReleaseImage(&pdst);cvReleaseImage(&tsimg);
225 | 
226 | }
227 | 


--------------------------------------------------------------------------------
/FDImage.h:
--------------------------------------------------------------------------------
 1 | #define USE_DOUBLE
 2 | 
 3 | #ifdef USE_DOUBLE
 4 | typedef double REAL;
 5 | #else
 6 | typedef float REAL;
 7 | #endif
 8 | 
 9 | 
10 | #include <opencv2/core/core.hpp>
11 | #include <opencv2/highgui/highgui.hpp>
12 | #include <opencv2/opencv.hpp>
13 | // 新版本写在下面文件中：
14 | #include <opencv2/nonfree/features2d.hpp>
15 | //#include "opencv2/features2d/features2d.hpp"
16 | #include<opencv2/legacy/legacy.hpp>
17 | 
18 | class FDImage
19 | {
20 | public:
21 | 	FDImage();
22 | 	~FDImage();
23 | 
24 | 	void Clear(void); 
25 | 	void SetSize(const CvSize size);
26 | 	FDImage& operator=(const FDImage& source);
27 | 	void Resize(FDImage &result,  REAL ratio) const;
28 | 	void Resize240x40(FDImage &result) const;
29 | 	void CutToLxL(FDImage &result, CvRect rect, int iL) const;
30 | 	void Copy(const FDImage& source);
31 | 	void Load(IplImage* img);
32 | 	void cleartmp();
33 | 
34 | public:
35 | 	int height; // height, or, number of rows of the image
36 | 	int width;  // width, or, number of columns of the image
37 | 	REAL** data;  // auxiliary pointers to accelerate the read/write of the image
38 | 	// no memory is really allocated, use memory in (buf)
39 | 	// data[i][j] is a pixel's gray value in (i)th row and (j)th column
40 | 	REAL* buf;    // pointer to a block of continuous memory containing the image
41 | 	int label;
42 | 	REAL variance;
43 | 
44 | //	SIFT
45 | 	//Mat 
46 | 
47 | 	IplImage* img;
48 | };
49 | 


--------------------------------------------------------------------------------
/NeuralNetWork.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/NeuralNetWork.cpp


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
  1 | 
  2 | 公众号MachineLN，邀请您扫码关注： 
  3 | 
  4 | ![image](http://upload-images.jianshu.io/upload_images/4618424-3ef1722341ba72d2?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240) 
  5 | 
  6 | 
  7 | **机器学习进阶系列：(下面内容在微信公众号同步)**
  8 | 
  9 | 1. [机器学习-1：MachineLN之三要素](http://blog.csdn.net/u014365862/article/details/78955063)
 10 | 
 11 | 2. [机器学习-2：MachineLN之模型评估](http://blog.csdn.net/u014365862/article/details/78959353)
 12 | 
 13 | 3. [机器学习-3：MachineLN之dl](http://blog.csdn.net/u014365862/article/details/78980142)
 14 | 
 15 | 4. [机器学习-4：DeepLN之CNN解析](http://blog.csdn.net/u014365862/article/details/78986089)
 16 | 
 17 | 5. [机器学习-5：DeepLN之CNN权重更新（笔记）](http://blog.csdn.net/u014365862/article/details/78959211)
 18 | 
 19 | 6. [机器学习-6：DeepLN之CNN源码](http://blog.csdn.net/u014365862/article/details/79010248)
 20 | 
 21 | 7. [机器学习-7：MachineLN之激活函数](http://blog.csdn.net/u014365862/article/details/79007801)
 22 | 
 23 | 8. [机器学习-8：DeepLN之BN](http://blog.csdn.net/u014365862/article/details/79019518)
 24 | 
 25 | 9. [机器学习-9：MachineLN之数据归一化](http://blog.csdn.net/u014365862/article/details/79031089)
 26 | 
 27 | 10. [机器学习-10：MachineLN之样本不均衡](http://blog.csdn.net/u014365862/article/details/79040390)
 28 | 
 29 | 11. [机器学习-11：MachineLN之过拟合](http://blog.csdn.net/u014365862/article/details/79057073) 
 30 | 
 31 | 12. [机器学习-12：MachineLN之优化算法](http://blog.csdn.net/u014365862/article/details/79070721)
 32 | 
 33 | 13. [机器学习-13：MachineLN之kNN](http://blog.csdn.net/u014365862/article/details/79091913)
 34 | 
 35 | 14. [机器学习-14：MachineLN之kNN源码](http://blog.csdn.net/u014365862/article/details/79101209)
 36 | 
 37 | 15. [](http://mp.blog.csdn.net/postedit/79135612)[机器学习-15：MachineLN之感知机](http://blog.csdn.net/u014365862/article/details/79135612)
 38 | 
 39 | 16. [机器学习-16：MachineLN之感知机源码](http://blog.csdn.net/u014365862/article/details/79135767)
 40 | 
 41 | 17. [机器学习-17：MachineLN之逻辑回归](http://blog.csdn.net/u014365862/article/details/79157777)
 42 | 
 43 | 18. [机器学习-18：MachineLN之逻辑回归源码](http://blog.csdn.net/u014365862/article/details/79157841)
 44 | 
 45 | 19. [机器学习-19：MachineLN之SVM（1）](http://blog.csdn.net/u014365862/article/details/79184858)
 46 | 
 47 | 20. [机器学习-20：MachineLN之SVM（2）](http://blog.csdn.net/u014365862/article/details/79202089)
 48 | 
 49 | 21. [机器学习-21：MachineLN之SVM源码](http://blog.csdn.net/u014365862/article/details/79224119)
 50 | 
 51 | 22. [机器学习-22：MachineLN之RL](http://blog.csdn.net/u014365862/article/details/79240997)
 52 | 
 53 | **人脸检测系列：**
 54 | 
 55 | 1. [人脸检测——AFLW准备人脸](http://blog.csdn.net/u014365862/article/details/74682464)
 56 | 
 57 | 2. [人脸检测——生成矫正人脸——cascade cnn的思想， 但是mtcnn的效果貌似更赞](http://blog.csdn.net/u014365862/article/details/74690865)
 58 | 
 59 | 3. [人脸检测——准备非人脸](http://blog.csdn.net/u014365862/article/details/74719498)
 60 | 
 61 | 4. [人脸检测——矫正人脸生成标签](http://blog.csdn.net/u014365862/article/details/74853099)
 62 | 
 63 | 5. [人脸检测——mtcnn思想，生成negative、positive、part样本。](http://blog.csdn.net/u014365862/article/details/78051411)
 64 | 
 65 | 6. [**人脸检测——滑动窗口篇（训练和实现）**](http://blog.csdn.net/u014365862/article/details/77816493)
 66 | 
 67 | 7. [**人脸检测——fcn**](http://blog.csdn.net/u014365862/article/details/78036382)
 68 | 
 69 | 8. [简单的人脸跟踪](http://blog.csdn.net/u014365862/article/details/77989896)
 70 | 
 71 | 9. [Face Detection(OpenCV) Using Hadoop Streaming API](http://blog.csdn.net/u014365862/article/details/78173740)
 72 | 
 73 | 10. [Face Recognition(face_recognition) Using Hadoop Streaming API](http://blog.csdn.net/u014365862/article/details/78175803)
 74 | 
 75 | 11. [非极大值抑制（Non-Maximum-Suppression）](http://blog.csdn.net/u014365862/article/details/53376516)
 76 | 
 77 | **OCR系列：**
 78 | 
 79 | **1. [tf20: CNN—识别字符验证码](http://blog.csdn.net/u014365862/article/details/53869816)** 
 80 | 
 81 | 2. [**身份证识别——生成身份证号和汉字**](http://blog.csdn.net/u014365862/article/details/78581949)
 82 | 
 83 | 3. [**tf21: 身份证识别——识别身份证号**](http://blog.csdn.net/u014365862/article/details/78582128)
 84 | 
 85 | 4. **[tf22: ocr识别——不定长数字串识别](http://blog.csdn.net/u014365862/article/details/78582417)**
 86 | 
 87 | **机器学习，深度学习系列:**
 88 | 
 89 | 1. [反向传播与它的直观理解](http://blog.csdn.net/u014365862/article/details/54728707)
 90 | 
 91 | 2. [卷积神经网络（CNN）：从原理到实现](http://blog.csdn.net/u014365862/article/details/54865609)
 92 | 
 93 | 3. [机器学习算法应用中常用技巧-1](http://blog.csdn.net/u014365862/article/details/54890040)
 94 | 
 95 | 4. [机器学习算法应用中常用技巧-2](http://blog.csdn.net/u014365862/article/details/54890046)
 96 | 
 97 | 5. [一个隐马尔科夫模型的应用实例：中文分词](http://blog.csdn.net/u014365862/article/details/54891582)
 98 | 
 99 | 6. [**Pandas处理csv表格**](http://blog.csdn.net/u014365862/article/details/54923429)
100 | 
101 | 7. [sklearn查看数据分布](http://blog.csdn.net/u014365862/article/details/54973495)
102 | 
103 | 8. [TensorFlow 聊天机器人](http://blog.csdn.net/u014365862/article/details/57518873)
104 | 
105 | 9. [YOLO](http://blog.csdn.net/u014365862/article/details/60321879)
106 | 
107 | 10. [感知机--模型与策略](http://blog.csdn.net/u014365862/article/details/61413859)
108 | 
109 | 11. [从 0 到 1 走进 Kaggle](http://blog.csdn.net/u014365862/article/details/72794198)
110 | 
111 | 12. [python调用Face++，玩坏了！](http://blog.csdn.net/u014365862/article/details/74149097)
112 | 
113 | 13. [人脸识别keras实现教程](http://blog.csdn.net/u014365862/article/details/74332028)
114 | 
115 | 14. [机器学习中的Bias(偏差)，Error(误差)，和Variance(方差)有什么区别和联系？](http://blog.csdn.net/u014365862/article/details/76360351)
116 | 
117 | 15. [CNN—pooling层的作用](http://blog.csdn.net/u014365862/article/details/77159143)
118 | 
119 | 16. [trick—Batch Normalization](http://blog.csdn.net/u014365862/article/details/77159778)
120 | 
121 | 17. [**tensorflow使用BN—Batch Normalization**](http://blog.csdn.net/u014365862/article/details/77188011)
122 | 
123 | 18. [trick—Data Augmentation](http://blog.csdn.net/u014365862/article/details/77193754)
124 | 
125 | 19. [CNN图图图](http://blog.csdn.net/u014365862/article/details/77367172)
126 | 
127 | 20. [为什么很多做人脸的Paper会最后加入一个Local Connected Conv？](http://blog.csdn.net/u014365862/article/details/77795902)
128 | 
129 | 21. [**Faster RCNN：RPN，anchor，sliding windows**](http://blog.csdn.net/u014365862/article/details/77887230)
130 | 
131 | 22. [**深度学习这些坑你都遇到过吗？**](http://blog.csdn.net/u014365862/article/details/77961624)
132 | 
133 | 23. [**image——Data Augmentation的代码**](http://blog.csdn.net/u014365862/article/details/78086604)
134 | 
135 | 24. [8种常见机器学习算法比较](http://blog.csdn.net/u014365862/article/details/52937983)
136 | 
137 | 25. [几种常见的激活函数](http://blog.csdn.net/u014365862/article/details/52710698)
138 | 
139 | 26. [**Building powerful image classification models using very little data**](http://blog.csdn.net/u014365862/article/details/78519629)
140 | 
141 | 27. [**机器学习模型训练时候tricks**](http://blog.csdn.net/u014365862/article/details/78519727)
142 | 
143 | 28. [OCR综述](https://handong1587.github.io/deep_learning/2015/10/09/ocr.html#handwritten-recognition)
144 | 
145 | 29. [一个有趣的周报](http://blog.csdn.net/u014365862/article/details/78757109)
146 | 
147 | 30. [根据已给字符数据，训练逻辑回归、随机森林、SVM，生成ROC和箱线图](http://blog.csdn.net/u014365862/article/details/78835541)
148 | 
149 | **图像处理系列：**
150 | 
151 | 1. [python下使用cv2.drawContours填充轮廓颜色](http://blog.csdn.net/u014365862/article/details/77720368)
152 | 
153 | 2. [imge stitching图像拼接stitching](http://blog.csdn.net/u014365862/article/details/53433048)
154 | 
155 | 3. [用python简单处理图片（1）：打开\显示\保存图像](http://blog.csdn.net/u014365862/article/details/52652256)
156 | 
157 | 4. [用python简单处理图片（2）：图像通道\几何变换\裁剪](http://blog.csdn.net/u014365862/article/details/52652273)
158 | 
159 | 5. [用python简单处理图片（3）：添加水印](http://blog.csdn.net/u014365862/article/details/52652296)
160 | 
161 | 6. [用python简单处理图片（4）：图像中的像素访问](http://blog.csdn.net/u014365862/article/details/52652300)
162 | 
163 | 7. [用python简单处理图片（5）：图像直方图](http://blog.csdn.net/u014365862/article/details/52652309)
164 | 
165 | 8. [**仿射变换，透视变换：二维坐标到二维坐标之间的线性变换，可用于landmark人脸矫正。**](http://blog.csdn.net/u014365862/article/details/78678872)
166 | 
167 | **代码整合系列：**
168 | 
169 | 1. [windows下C++如何调用matlab程序](http://blog.csdn.net/u014365862/article/details/77480325)
170 | 
171 | 2. [ubuntu下C++如何调用matlab程序](http://blog.csdn.net/u014365862/article/details/77529096)
172 | 
173 | 3. [matlab使用TCP/IP Server Sockets](http://blog.csdn.net/u014365862/article/details/77745476)
174 | 
175 | 4. [ubuntu下C++如何调用python程序，gdb调试C++代码](http://blog.csdn.net/u014365862/article/details/77864743)
176 | 
177 | 5. [How to pass an array from C++ to an embedded python](http://blog.csdn.net/u014365862/article/details/77891487)
178 | 
179 | 6. [如何使用Python为Hadoop编写一个简单的MapReduce程序](http://blog.csdn.net/u014365862/article/details/78169554)
180 | 
181 | 7. [图像的遍历](http://blog.csdn.net/u014365862/article/details/53513710)
182 | 
183 | 8. [**ubuntu下CMake编译生成动态库和静态库，以OpenTLD为例。**](http://blog.csdn.net/u014365862/article/details/78663269)
184 | 
185 | 9. [**ubuntu下make编译生成动态库，然后python调用cpp。**](http://blog.csdn.net/u014365862/article/details/78675033)
186 | 
187 | **数据结构和算法系列：**
188 | 
189 | 1. [堆排序](http://blog.csdn.net/u014365862/article/details/78200711)
190 | 
191 | 2. [red and black (深度优先搜索算法dfs)](http://blog.csdn.net/u014365862/article/details/48781603)
192 | 
193 | 3. [深度优先搜索算法](http://blog.csdn.net/u014365862/article/details/48729681)
194 | 
195 | 4. [qsort原理和实现](http://blog.csdn.net/u014365862/article/details/48688457)
196 | 
197 | 5. [stack实现queue ; list实现stack](http://blog.csdn.net/u014365862/article/details/48594323)
198 | 
199 | 6. [另一种斐波那契数列](http://blog.csdn.net/u014365862/article/details/48573545)
200 | 
201 | 7. [堆和栈的区别(个人感觉挺不错的)](http://blog.csdn.net/u014365862/article/details/49159499)
202 | 
203 | 8. [排序方法比较](http://blog.csdn.net/u014365862/article/details/52502824)
204 | 
205 | 9. [漫画 ：什么是红黑树？](https://mp.weixin.qq.com/s/JJVbi7kqDpLUuh696J7oLg)
206 | 
207 | 10. [牛客网刷题](https://www.nowcoder.com/activity/oj)
208 | 
209 | 11. [莫烦python 666](https://morvanzhou.github.io/)
210 | 
211 | **kinect 系列：**
212 | 
213 | 1. [Kinect v2.0原理介绍之一：硬件结构](http://blog.csdn.net/u014365862/article/details/46713807)
214 | 
215 | 2. [Kinect v2.0原理介绍之二：6种数据源](http://blog.csdn.net/u014365862/article/details/46849253)
216 | 
217 | 3. [Kinect v2.0原理介绍之三：骨骼跟踪的原理](http://blog.csdn.net/u014365862/article/details/46849309)
218 | 
219 | 4. [Kinect v2.0原理介绍之四：人脸跟踪探讨](http://blog.csdn.net/u014365862/article/details/46849357)
220 | 
221 | 5. [Kinect v2.0原理介绍之五：只检测离kinect最近的人脸](http://blog.csdn.net/u014365862/article/details/47809401)
222 | 
223 | 6. [Kinect v2.0原理介绍之六：Kinect深度图与彩色图的坐标校准](http://blog.csdn.net/u014365862/article/details/48212085)
224 | 
225 | 7. [Kinect v2.0原理介绍之七：彩色帧获取](http://blog.csdn.net/u014365862/article/details/48212377)
226 | 
227 | 8. [Kinect v2.0原理介绍之八：高清面部帧(1) FACS 介绍](http://blog.csdn.net/u014365862/article/details/48212631)
228 | 
229 | 9. [Kinect v2.0原理介绍之九：高清面部帧(2) 面部特征对齐](http://blog.csdn.net/u014365862/article/details/48212757)
230 | 
231 | 10. [Kinect v2.0原理介绍之十：获取高清面部帧的AU单元特征保存到文件](http://blog.csdn.net/u014365862/article/details/48780361)
232 | 
233 | 11. [kinect v2.0原理介绍之十一：录制视频](http://blog.csdn.net/u014365862/article/details/77929405)
234 | 
235 | 12. [Kinect v2.0原理介绍之十二：音频获取](http://blog.csdn.net/u014365862/article/details/49204931)
236 | 
237 | 13. [Kinect v2.0原理介绍之十三：面部帧获取](http://blog.csdn.net/u014365862/article/details/50434088)
238 | 
239 | 14. [Kinect for Windows V2和V1对比开发___彩色数据获取并用OpenCV2.4.10显示](http://blog.csdn.net/u014365862/article/details/48948861)
240 | 
241 | 15. [Kinect for Windows V2和V1对比开发___骨骼数据获取并用OpenCV2.4.10显示](http://blog.csdn.net/u014365862/article/details/48949055)
242 | 
243 | 16. [用kinect录视频库](http://blog.csdn.net/u014365862/article/details/48946543)
244 | 
245 | **tensorflow系列：**
246 | 
247 | 1. [](http://blog.csdn.net/u014365862/article/details/78422315)[Ubuntu 16.04 安装 Tensorflow(GPU支持)](http://blog.csdn.net/u014365862/article/details/53868411)
248 | 
249 | 2. [使用Python实现神经网络](http://blog.csdn.net/u014365862/article/details/53868414)
250 | 
251 | 3. [tf1: nn实现评论分类](http://blog.csdn.net/u014365862/article/details/53868418)
252 | 
253 | 4. [tf2: nn和cnn实现评论分类](http://blog.csdn.net/u014365862/article/details/53868422)
254 | 
255 | 5. [tf3: RNN—mnist识别](http://blog.csdn.net/u014365862/article/details/53868425)
256 | 
257 | 6. [tf4: CNN—mnist识别](http://blog.csdn.net/u014365862/article/details/53868430)
258 | 
259 | 7\.  [tf5: Deep Q Network—AI游戏](http://blog.csdn.net/u014365862/article/details/53868436)
260 | 
261 | 8. [tf6: autoencoder—WiFi指纹的室内定位](http://blog.csdn.net/u014365862/article/details/53868533)
262 | 
263 | 9. [tf7: RNN—古诗词](http://blog.csdn.net/u014365862/article/details/53868544)
264 | 
265 | 10. [tf8:RNN—生成音乐](http://blog.csdn.net/u014365862/article/details/53868549)
266 | 
267 | 11. [tf9: PixelCNN](http://blog.csdn.net/u014365862/article/details/53868557)
268 | 
269 | 12. [tf10: 谷歌Deep Dream](http://blog.csdn.net/u014365862/article/details/53868560)
270 | 
271 | 13. [tf11: retrain谷歌Inception模型](http://blog.csdn.net/u014365862/article/details/53868568)
272 | 
273 | 14. [tf12: 判断男声女声](http://blog.csdn.net/u014365862/article/details/54600398)
274 | 
275 | 15. [tf13: 简单聊天机器人](http://blog.csdn.net/u014365862/article/details/53869660)
276 | 
277 | 16. [tf14: 黑白图像上色](http://blog.csdn.net/u014365862/article/details/53869682)
278 | 
279 | 17. [tf15: 中文语音识别](http://blog.csdn.net/u014365862/article/details/53869701)
280 | 
281 | 18. [tf16: 脸部特征识别性别和年龄](http://blog.csdn.net/u014365862/article/details/53869712)
282 | 
283 | 19. [tf17: “声音大挪移”](http://blog.csdn.net/u014365862/article/details/53869724)
284 | 
285 | 20. [tf18: 根据姓名判断性别](http://blog.csdn.net/u014365862/article/details/53869732)
286 | 
287 | 21\.  [tf19: 预测铁路客运量](http://blog.csdn.net/u014365862/article/details/53869802)
288 | 
289 | 22. [**tf20: CNN—识别字符验证码**](http://blog.csdn.net/u014365862/article/details/53869816)
290 | 
291 | 23. [tf21: 身份证识别——识别身份证号](http://blog.csdn.net/u014365862/article/details/78582128)
292 | 
293 | 24. [](http://blog.csdn.net/u014365862/article/details/78582417)[tf22: ocr识别——不定长数字串识别](http://blog.csdn.net/u014365862/article/details/78582417)
294 | 
295 | 25. [tf23: “恶作剧” --人脸检测](http://blog.csdn.net/u014365862/article/details/53978811)
296 | 
297 | 26. [tf24: GANs—生成明星脸](http://blog.csdn.net/u014365862/article/details/54380277)
298 | 
299 | 27. [](http://blog.csdn.net/u014365862/article/details/54706771)[tf25: 使用深度学习做阅读理解+完形填空](http://blog.csdn.net/u014365862/article/details/54428325)
300 | 
301 | 28. [tf26: AI操盘手](http://blog.csdn.net/u014365862/article/details/54706771)
302 | 
303 | 29. [tensorflow_cookbook--preface](http://blog.csdn.net/u014365862/article/details/70837573)
304 | 
305 | 30. [01 TensorFlow入门（1）](http://blog.csdn.net/u014365862/article/details/70837638)
306 | 
307 | 31. [01 TensorFlow入门（2）](http://blog.csdn.net/u014365862/article/details/70849334)
308 | 
309 | 32. [02 The TensorFlow Way（1）](http://blog.csdn.net/u014365862/article/details/70884624)
310 | 
311 | 33. [02 The TensorFlow Way（2）](http://blog.csdn.net/u014365862/article/details/70887213)
312 | 
313 | 34. [02 The TensorFlow Way（3）](http://blog.csdn.net/u014365862/article/details/71038528)
314 | 
315 | 35. [03 Linear Regression](http://blog.csdn.net/u014365862/article/details/71064855)
316 | 
317 | 36. [04 Support Vector Machines](http://blog.csdn.net/u014365862/article/details/71078010)
318 | 
319 | 37. [tf API 研读1：tf.nn，tf.layers， tf.contrib概述](http://blog.csdn.net/u014365862/article/details/77833481)
320 | 
321 | 38. [tf API 研读2：math](http://blog.csdn.net/u014365862/article/details/77847410)
322 | 
323 | 39. [tensorflow中的上采样(unpool)和反卷积(conv2d_transpose)](http://blog.csdn.net/u014365862/article/details/77936259)
324 | 
325 | 40. [tf API 研读3：Building Graphs](http://blog.csdn.net/u014365862/article/details/77944301)
326 | 
327 | 41. [tf API 研读4：Inputs and Readers](http://blog.csdn.net/u014365862/article/details/77946268)
328 | 
329 | 42. [](http://blog.csdn.net/u014365862/article/details/77967231)[tf API 研读5：Data IO](http://blog.csdn.net/u014365862/article/details/77967231)
330 | 
331 | 43. [tf API 研读6：Running Graphs](http://blog.csdn.net/u014365862/article/details/77967995)
332 | 
333 | 44. [**tf.contrib.rnn.static_rnn与tf.nn.dynamic_rnn区别**](http://blog.csdn.net/u014365862/article/details/78238807)
334 | 
335 | 45. [**Tensorflow使用的预训练的resnet_v2_50，resnet_v2_101，resnet_v2_152等模型预测，训练**](http://blog.csdn.net/u014365862/article/details/78272811)
336 | 
337 | 46. [**tensorflow下设置使用某一块GPU、多GPU、CPU的情况**](http://blog.csdn.net/u014365862/article/details/78292475)
338 | 
339 | 47. [**工业器件检测和识别**](http://blog.csdn.net/u014365862/article/details/78359194)
340 | 
341 | 48. [**将tf训练的权重保存为CKPT,PB ,CKPT 转换成 PB格式。并将权重固化到图里面,并使用该模型进行预测**](http://blog.csdn.net/u014365862/article/details/78404980)
342 | 
343 | 49. **[tensorsor快速获取所有变量，和快速计算L2范数](http://blog.csdn.net/u014365862/article/details/78422315)**
344 | 
345 | 50. [**cnn+rnn+attention**](http://blog.csdn.net/u014365862/article/details/78495870)
346 | 
347 | 51. [Tensorflow实战学习笔记](https://github.com/MachineLP/Tensorflow-)
348 | 
349 | 52. [tf27: Deep Dream—应用到视频](http://blog.csdn.net/u014365862/article/details/53869830)
350 | 
351 | 53. [tf28: 手写汉字识别](http://blog.csdn.net/u014365862/article/details/53869837)
352 | 
353 | 54. [tf29: 使用tensorboard可视化inception_v4](http://blog.csdn.net/u014365862/article/details/79115556)
354 | 
355 | 55. [tf30: center loss及其mnist上的应用](http://blog.csdn.net/u014365862/article/details/79184966)
356 | 
357 | 56. [tf31: keras的LSTM腾讯人数在线预测](http://blog.csdn.net/u014365862/article/details/79186993)
358 | 
359 | 57. [tf32: 一个简单的cnn模型：人脸特征点训练](http://blog.csdn.net/u014365862/article/details/79187157)
360 | 
361 | 58. [tf33: 图片降噪：卷积自编码](http://blog.csdn.net/u014365862/article/details/79246179)
362 | 
363 | **C++系列：**
364 | 
365 | 1. [c++ primer之const限定符](http://blog.csdn.net/u014365862/article/details/46848613)
366 | 
367 | 2. [c++primer之auto类型说明符](http://blog.csdn.net/u014365862/article/details/46849697)
368 | 
369 | 3. [c++primer之预处理器](http://blog.csdn.net/u014365862/article/details/46853869)
370 | 
371 | 4. [c++primer之string](http://blog.csdn.net/u014365862/article/details/46860037)
372 | 
373 | 5. [c++primer之vector](http://blog.csdn.net/u014365862/article/details/46885087)
374 | 
375 | 6. [c++primer之多维数组](http://blog.csdn.net/u014365862/article/details/46933199)
376 | 
377 | 7. [c++primer之范围for循环](http://blog.csdn.net/u014365862/article/details/47706255)
378 | 
379 | 8. [c++primer之运算符优先级表](http://blog.csdn.net/u014365862/article/details/47706423)
380 | 
381 | 9. [c++primer之try语句块和异常处理](http://blog.csdn.net/u014365862/article/details/47707669)
382 | 
383 | 10. [c++primer之函数(函数基础和参数传递)](http://blog.csdn.net/u014365862/article/details/47783193)
384 | 
385 | 11. [c++primer之函数(返回类型和return语句)](http://blog.csdn.net/u014365862/article/details/47808711)
386 | 
387 | 12. [c++primer之函数重载](http://blog.csdn.net/u014365862/article/details/47834667)
388 | 
389 | 13. [c++重写卷积网络的前向计算过程，完美复现theano的测试结果](http://blog.csdn.net/u014365862/article/details/48010697)
390 | 
391 | 14. [c++ primer之类](http://blog.csdn.net/u014365862/article/details/48165685)
392 | 
393 | 15. [c++primer之类（构造函数再探）](http://blog.csdn.net/u014365862/article/details/48198595)
394 | 
395 | 16. [c++primer之类（类的静态成员）](http://blog.csdn.net/u014365862/article/details/48199161)
396 | 
397 | 17. [c++primer之顺序容器（容器库概览）](http://blog.csdn.net/u014365862/article/details/48209805)
398 | 
399 | 18. [c++primer之顺序容器（添加元素）](http://blog.csdn.net/u014365862/article/details/48226673)
400 | 
401 | 19. [c++primer之顺序容器（访问元素）](http://blog.csdn.net/u014365862/article/details/48230053)
402 | 
403 | **OpenCV系列：**
404 | 
405 | 1. [自己训练SVM分类器，进行HOG行人检测。](http://blog.csdn.net/u014365862/article/details/53243604)
406 | 
407 | 2. [opencv-haar-classifier-training](http://blog.csdn.net/u014365862/article/details/53096367)
408 | 
409 | 3. [vehicleDectection with Haar Cascades](http://blog.csdn.net/u014365862/article/details/53087675)
410 | 
411 | 4. [LaneDetection](http://blog.csdn.net/u014365862/article/details/53083143)
412 | 
413 | 5. [OpenCV学习笔记大集锦](http://blog.csdn.net/u014365862/article/details/53063627)
414 | 
415 | 6. [Why always OpenCV Error: Assertion failed (elements_read == 1) in unknown function ?](http://blog.csdn.net/u014365862/article/details/53000619)
416 | 
417 | 7. [目标检测之训练opencv自带的分类器（opencv_haartraining 或 opencv_traincascade）](http://blog.csdn.net/u014365862/article/details/52997019)
418 | 
419 | 8. [车牌识别 之 字符分割](http://blog.csdn.net/u014365862/article/details/52672747)
420 | 
421 | 9. **[仿射变换，透视变换：二维坐标到二维坐标之间的线性变换，可用于landmark人脸矫正。](http://blog.csdn.net/u014365862/article/details/78678872)**
422 | 
423 | 10. [opencv实现抠图（单一背景），替换背景图](http://blog.csdn.net/u014365862/article/details/78863756)
424 | 
425 | **python系列（**web开发、多线程等**）：** 
426 | 
427 | 1. [**flask的web开发，用于机器学习（主要还是DL）模型的简单演示。**](http://blog.csdn.net/u014365862/article/details/78818334)
428 | 
429 | 2. **[python多线程，获取多线程的返回值](http://blog.csdn.net/u014365862/article/details/78835348)**
430 | 
431 | 3. [文件中字的统计及创建字典](http://blog.csdn.net/u014365862/article/details/78914151)
432 | 
433 | **其他:**
434 | 
435 | 1. [MAC平台下Xcode配置使用OpenCV的具体方法 (2016最新)](http://blog.csdn.net/u014365862/article/details/53067565)
436 | 
437 | 2. [**python下如何安装.whl包？**](http://blog.csdn.net/u014365862/article/details/51817390)
438 | 
439 | 3. [给中国学生的第三封信：成功、自信、快乐](http://blog.csdn.net/u014365862/article/details/47972321)
440 | 
441 | 4. [自己-社会-机器学习](http://blog.csdn.net/u014365862/article/details/48604145)
442 | 
443 | 5. [不执著才叫看破，不完美才叫人生。](http://blog.csdn.net/u014365862/article/details/49079047)
444 | 
445 | 6. [PCANet的C++代码——详细注释版](http://blog.csdn.net/u014365862/article/details/51213280)
446 | 
447 | 7. [责任与担当](http://blog.csdn.net/u014365862/article/details/51841590)
448 | 
449 | 8. [好走的都是下坡路](http://blog.csdn.net/u014365862/article/details/53244402)
450 | 
451 | 9. [一些零碎的语言，却触动到内心深处。](http://blog.csdn.net/u014365862/article/details/53186012)
452 | 
453 | 10. [用一个脚本学习 python](http://blog.csdn.net/u014365862/article/details/54428373)
454 | 
455 | 11. [一个有趣的周报](http://blog.csdn.net/u014365862/article/details/78757109)
456 | 


--------------------------------------------------------------------------------
/cnn.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/cnn.cpp


--------------------------------------------------------------------------------
/cnn.h:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/cnn.h


--------------------------------------------------------------------------------
/data/.DS_Store:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/data/.DS_Store


--------------------------------------------------------------------------------
/data/cascadeClassifier.txt:
--------------------------------------------------------------------------------
 1 | 8
 2 | 4 0.507175
 3 | 0 8 20 30 -0.359593 0.168139 0.128429 0.0500019 0.278004 0.112702 -0.138898 -0.00123301 0.201154 0.307577 0.245603 -0.111773 -0.0109768 0.203608 -0.0564578 -0.0215202 0.095214 0.208032 0.319649 -0.0951572 -0.19111 -0.185246 0.0292464 0.0378044 0.0417406 -0.0582106 0.123968 0.306425 0.222741 -0.0412855 0.212866 0.045826 0.0234087
 4 | 20 0 40 20 -0.0993183 0.0154231 -0.170813 0.266289 -0.0814961 -0.0257754 0.0795825 0.320839 -0.218228 0.124429 0.0462424 -0.0713423 -0.00829158 0.0788605 -0.0338962 0.0116684 -0.00445853 -0.108481 -0.0715571 0.178786 0.231159 -0.0308938 -0.0289002 0.0880318 0.186511 0.300366 0.247873 -0.232104 -0.364099 -0.362693 -0.178358 0.176247 -0.072362
 5 | 0 0 40 20 -0.181522 0.0372686 0.0291435 0.0285497 0.0876193 0.0102889 -0.0299387 0.0314655 0.0685195 -0.0314048 0.0611647 0.117547 -0.0368222 -0.209555 0.144075 0.0951549 0.0655656 0.330661 0.351931 -0.193426 -0.359123 0.0452908 0.244819 -0.175597 -0.359123 -0.136961 -0.165013 0.252834 0.304812 0.099339 0.0465817 0.0682139 0.0136863
 6 | 8 8 20 30 -0.353141 0.259153 0.286968 -0.149303 -0.0322963 0.0456567 -0.0681416 0.036441 0.232315 0.0984322 0.144887 -0.0498762 0.268708 0.209378 -0.0677844 -0.161367 0.219082 0.0835215 6.04756e-006 0.249265 0.163714 -0.0742528 -0.0685105 0.229229 0.0491831 0.309446 0.244516 -0.0677036 -0.204498 0.0218611 0.0869947 0.0356375 -0.173034
 7 | 
 8 | 4 0.489683
 9 | 4 8 30 20 -0.286726 -0.0237679 -0.139003 0.0357271 0.224666 0.204132 -0.0967802 0.0433742 0.126365 0.16342 0.149294 -0.13685 -0.297535 0.036858 0.214934 0.202266 0.0283493 0.0947439 0.155522 -0.0856717 -0.134323 0.154863 0.232629 -0.110399 -0.102889 -0.302226 -0.301861 0.18525 0.302285 0.161989 0.060991 0.00569456 0.155812
10 | 8 4 12 18 -0.344338 0.0598853 -0.0765459 0.349926 0.304556 0.248375 0.173918 -0.0872134 -0.126546 0.232795 0.0796251 0.0918951 -0.00422765 0.00689119 -0.0779652 0.184134 0.00562018 -0.00496651 0.143342 -0.0282139 0.187536 -0.0669503 -0.20637 -0.0205052 0.233764 0.205656 0.195528 -0.305215 -0.0066716 0.145348 -0.0445885 -0.249013 0.0674026
11 | 20 20 20 20 -0.131017 -0.0113266 -0.013239 0.185442 0.12111 0.0110295 0.127629 -0.0204003 -0.0269157 0.202446 0.151924 -0.128455 -0.310094 -0.139432 -0.0308167 0.0304799 -0.104737 -0.114983 -0.0629965 0.16078 0.193244 -0.0488536 -0.116067 0.0743813 0.170445 0.30072 0.26222 -0.136326 -0.268842 -0.350017 -0.2778 0.300875 0.160353
12 | 16 0 12 24 -0.328614 -0.144955 -0.197597 0.12664 -0.0965862 -0.212797 -0.0521935 0.185381 -0.227959 0.106887 -0.110054 0.0676273 0.0767115 0.358213 0.12979 0.0364869 -0.213069 0.30908 0.251531 -0.104918 -0.00505632 0.145153 0.0351957 0.0880433 0.119813 -0.0609272 0.0812291 0.259965 0.316298 -0.057757 0.122057 -0.0898008 0.131522
13 | 
14 | 4 0.488712
15 | 4 0 40 20 -0.0919957 -0.219091 -0.276803 0.09001 0.120394 0.125377 0.0420575 0.154562 0.150409 0.0800062 0.128578 -0.296283 -0.352692 -0.0133781 0.146715 0.0762344 0.0188867 0.0490024 0.0645908 -0.195127 -0.233938 0.117011 0.167533 -0.0383378 -0.0635173 -0.352692 -0.350928 0.141627 0.159481 0.162023 0.140972 0.0896059 0.1151
16 | 28 0 24 12 -0.151764 -0.14136 -0.0345763 0.359054 0.269019 -0.0511534 0.0862612 0.197281 -0.0235772 0.340465 0.357769 -0.321892 -0.0880085 0.0749356 -0.10016 -0.282274 -0.109026 -0.0680141 -0.0500643 0.164303 0.097643 0.141432 0.0986346 -0.132917 -0.124829 0.119809 0.0645598 -0.175312 -0.175548 -0.0496006 -0.143953 0.146357 0.022847
17 | 0 20 20 20 -0.223475 0.132541 0.174064 -0.125524 -0.320947 -0.0667298 0.041751 -0.00992267 -0.098804 0.0335989 -0.00389866 0.201758 0.143897 -0.0234615 0.0797261 0.153152 0.0798391 0.301835 0.265678 -0.0920869 -0.169792 0.177971 0.274013 -0.267165 -0.347481 -0.0454768 -0.112282 0.108272 0.214742 0.244338 0.0993439 -0.116285 -0.0681153
18 | 0 0 6 24 0.00189935 0.243988 0.277686 -0.267446 -0.283723 -0.163387 -0.25689 -0.021889 0.228091 -0.235 -0.241646 0.35304 0.117759 0.00805924 0.0896609 -0.0878764 0.0510719 -0.239205 0.0137837 0.199488 -0.034473 -0.0426234 0.0890696 -0.258274 -0.259156 0.0393429 0.0187748 0.0838972 -0.0386316 -0.102853 -0.130301 0.0304256 0.0661349
19 | 
20 | 11 0.500562
21 | 16 4 30 20 -0.346605 0.2464 0.23436 -0.229307 -0.321937 0.100754 0.175418 0.0742107 -0.0247097 -0.224333 -0.148268 0.15967 0.244843 0.129013 0.0918291 0.0604785 0.230322 -0.133739 -0.159559 0.229371 0.202754 0.114368 0.156968 0.166338 0.117706 0.15639 0.11855 -0.151561 -0.174078 0.0232208 0.0161727 0.0500643 -0.0107622
22 | 0 8 18 12 -0.248362 0.213957 0.160232 -0.0272569 0.00610056 0.0839336 0.0184162 -0.113228 -0.0942647 0.0719929 -0.107868 0.326481 0.355832 0.0289348 -0.0635148 -0.0315413 0.0758271 0.162391 0.259328 -0.190869 -0.288445 0.0451809 0.235615 -0.139217 -0.106576 0.118038 0.137365 0.0922976 -0.102715 -0.351225 -0.14937 0.196297 0.165179
23 | 20 8 30 20 -0.221091 0.159224 0.058722 0.0413889 0.189929 0.346941 0.146188 -0.1873 -0.244211 0.154293 0.0717609 -0.0974934 -0.0975289 -0.0179406 -0.0890791 0.113747 0.0290597 -0.17542 -0.151972 0.138323 0.307434 0.0408972 -0.113057 -0.0223947 0.248809 0.223429 0.250444 -0.103372 -0.269215 -0.25212 -0.0556975 0.240091 0.0602941
24 | 12 0 12 18 -0.350725 -0.0769412 -0.0557404 0.287073 0.0230446 0.00823636 0.147689 -0.053196 -0.308121 -0.234436 -0.303285 0.34929 0.245264 -0.0122943 -0.0546945 0.147664 0.00735194 0.19598 0.262152 0.00271067 0.0309456 -0.00561812 0.153834 0.051203 0.0879683 0.116867 0.0298137 0.160962 0.203412 0.0362125 -0.101631 0.199986 0.159704
25 | 0 28 12 24 -0.15 0.237497 0.107801 -0.144548 -0.210693 0.0850036 -0.00181229 -0.121797 -0.271254 -0.226178 -0.215315 0.297428 0.244004 0.0494057 0.12366 0.0395337 0.118261 0.13019 0.197856 -0.0074547 -0.0999798 0.151337 0.349295 -0.302844 -0.298295 -0.122923 -0.124745 0.0707806 0.0524016 0.136864 0.0919146 -0.0253115 -0.0409925
26 | 0 4 20 40 -0.370278 -0.0216801 0.166216 0.0873784 0.0916601 0.101133 0.215428 0.0615004 0.300593 0.164222 0.0857375 0.00538473 0.0890945 0.214797 0.111006 0.0989723 0.105805 0.0747669 -1.81878e-005 -0.0485214 -0.16683 -0.0942571 -0.0229704 0.370278 0.317976 0.109078 0.104615 -0.0492959 -0.2045 0.266313 0.305859 0.0509976 -0.131783
27 | 16 4 12 12 -0.0774259 0.208479 0.258217 -0.0420312 -0.126201 -0.0364822 0.0414582 -0.0659377 -0.105292 0.0202322 -0.00944028 0.16355 0.241801 -0.241168 -0.184584 0.146428 0.0677712 0.255377 0.241228 -0.352766 -0.124612 -0.0747156 -0.267036 0.0179486 0.0838067 0.168874 0.0477519 -0.308906 0.0105808 0.188873 -0.0848417 -0.337827 0.0499056
28 | 28 0 12 12 -0.0868713 -0.328849 -0.187516 0.117428 0.117018 0.164536 0.0763099 0.179433 -0.0931382 0.165756 0.15117 -0.202572 -0.130174 0.351021 0.101215 -0.285945 -0.258549 -0.220968 -0.103407 0.256994 0.315081 -0.161109 -0.111844 0.111912 0.0103489 0.129947 0.0898256 -0.0662535 0.0673252 -0.158056 -0.0910681 0.0391042 0.052294
29 | 28 28 12 12 0.0188162 -0.152734 0.0293583 0.206112 -0.0867642 0.12782 -0.0535637 -0.0677265 0.0399981 0.159396 0.276941 -0.151782 -0.313572 -0.175949 -0.106761 -0.0652095 -0.0708562 0.162909 0.117826 0.081095 -0.125885 -0.232332 -0.346837 0.130742 0.230689 0.133789 0.162033 -0.0781883 -0.107419 -0.292073 -0.290221 0.244905 0.127592
30 | 8 12 12 24 -0.375811 -0.0174387 -0.0177869 -0.135922 0.00903038 -0.0196359 -0.0716177 0.134419 0.375811 -0.0210086 0.0580564 -0.157925 -0.0262162 0.375811 0.273515 -0.075196 0.0536528 0.149496 0.094582 0.189421 0.176977 0.0530718 -0.0146651 0.281083 0.0982853 0.231982 0.246177 0.106733 0.0180672 0.0936492 0.278503 0.060474 0.052729
31 | 0 0 24 6 -0.0273608 0.0821286 -0.0446033 -0.147537 -0.146453 -0.329566 -0.0123025 0.128055 0.348904 0.234769 0.127543 0.081653 0.348904 -0.0447911 -0.119891 0.0443899 -0.303728 -0.0598588 -0.160024 -0.0938144 -0.120388 0.108828 0.308678 -0.321429 0.0375206 0.0696869 0.00224067 0.0284993 -0.295515 0.0350871 0.0969523 -0.0770352 -0.0779225
32 | 
33 | 12 0.497427
34 | 4 0 40 20 -0.271064 -0.129554 -0.202714 0.268208 0.280923 0.10775 0.0313935 0.106078 0.0736866 0.213506 0.269907 -0.183969 -0.324967 -0.0436285 0.147041 0.0620238 -0.0326638 0.162631 0.182594 -0.0376046 -0.120092 -0.00224906 0.128268 -0.0312025 -0.0964052 -0.25713 -0.24701 0.244375 0.281112 0.0887359 0.0141938 0.0723361 0.103742
35 | 20 16 20 20 -0.324734 0.294547 0.216558 0.237114 0.208047 0.159193 0.107366 -0.109301 -0.232769 0.128086 0.00832281 0.0451656 -0.172478 -0.0538625 0.0139115 0.156917 -0.184695 0.0384542 0.112051 0.169376 0.14968 0.06298 0.0852978 -0.0444691 0.0490919 0.253779 0.28185 -0.130237 -0.22867 -0.217894 -0.114185 0.22854 0.178315
36 | 8 4 12 24 -0.356361 -0.0634368 -0.0791345 0.182128 0.143773 0.109568 0.198892 -0.0700947 -0.0220684 0.0489699 -0.0412777 0.178228 0.145984 0.135818 0.0621221 0.325761 0.158068 0.103716 0.210801 -0.286355 0.0471016 -0.00509897 -0.265688 -0.0442144 0.335304 0.111512 -0.0256076 -0.219382 0.102637 0.29747 -0.115548 -0.151449 0.0995363
37 | 0 8 30 20 -0.178551 0.0913792 0.117326 -0.0373887 -0.0271932 0.0309849 0.0487832 -0.0897086 -0.085013 0.355742 0.310352 -0.0785361 -0.114333 -0.0844645 -0.14219 0.230666 0.140521 0.247527 0.240104 -0.15928 -0.253864 0.166242 0.25774 -0.192574 -0.321598 -0.16801 -0.136185 0.0742989 0.21005 0.164693 -0.0351495 -0.0185845 0.0629479
38 | 8 0 12 18 -0.288669 -0.25146 -0.209683 0.223017 0.133422 0.0312764 0.263863 -0.102652 0.0124145 -0.11614 -0.190595 0.171427 0.0467508 -0.0750314 -0.0387213 0.107554 -0.0640026 0.00753012 -0.0269267 0.35508 0.306282 0.0890053 0.108515 -0.0663318 -0.00027945 0.35508 0.295529 -0.236015 -0.0822912 0.0288799 -0.0937358 -0.0242157 0.12323
39 | 16 28 12 24 -0.189673 -0.29189 -0.0709571 0.350059 0.271272 0.0619018 -0.147983 0.195782 0.0767337 0.0801159 0.109946 -0.0103297 -0.154737 -0.0662952 0.0642548 -0.0554374 0.036993 -0.135306 -0.111575 0.0365612 0.0231982 -0.0141758 -0.0214738 0.0976744 0.183695 0.221105 0.111125 -0.0424265 -0.164195 -0.289163 -0.345932 0.34708 0.227497
40 | 0 16 24 12 0.0268079 0.0396889 0.114492 -0.0266056 -0.265453 -0.039721 0.13553 -0.0936007 -0.121478 -0.124589 -0.151461 0.216243 0.167935 -0.355688 -0.355688 0.15709 0.236838 0.24446 0.126471 -0.110926 -0.206321 -0.00270478 0.221975 -0.112806 -0.355688 0.210806 0.14784 -0.0267661 0.0314845 0.0614428 -0.0108984 -0.102121 -0.104047
41 | 12 0 40 10 -0.377735 0.23824 0.261313 0.0877716 -0.0797137 -0.00835802 0.293576 0.0652553 -0.0257275 0.131098 0.344153 -0.143469 -0.0533644 0.0575939 0.117823 0.0746699 0.262752 -0.0106354 -0.0984038 0.198224 0.290796 0.253971 0.029988 0.198831 -0.0533535 0.00948482 0.0740748 0.0935356 0.0918893 0.220487 0.163582 0.047849 0.150098
42 | 28 8 12 12 -0.126485 -0.22484 0.100564 0.203933 0.326273 0.0414935 -0.059262 0.0432824 -0.0261086 0.207407 0.260046 -0.163084 0.0798698 -0.191112 -0.183039 0.148653 -0.0273799 -0.280427 -0.202682 0.148453 0.0802211 0.33632 0.194253 -0.150723 -0.0909467 0.0085385 -0.00364751 -0.303363 -0.138208 -0.0411324 -0.10233 0.200769 0.165516
43 | 16 0 6 24 0.080976 0.0359464 0.1401 0.174824 -0.132536 -0.119003 -0.263483 -0.0474807 -0.105269 -0.177099 -0.265302 0.136726 0.0461379 -0.098618 -0.205534 0.073757 -0.103215 -0.25305 -0.278213 0.182705 0.0761604 0.162743 0.00547431 0.0530362 -0.0108112 0.0672201 0.162526 -0.197403 -0.209045 0.358267 0.0339476 -0.269885 -0.316606
44 | 0 28 12 24 -0.172546 0.186628 0.122718 -0.128761 -0.199862 0.0501401 -0.112258 0.0697911 0.00978351 -0.0223731 -0.21416 0.356319 0.150417 -0.0762529 -0.0363487 0.145771 0.163307 0.219491 0.175551 -0.0267573 -0.00462231 0.0314989 0.324988 -0.356319 -0.312037 -0.260772 -0.0933571 0.0149835 -0.0535189 0.182889 0.231 -0.0438307 0.0551526
45 | 32 16 24 6 -0.046273 -0.0760343 0.203314 0.204724 0.186031 -0.0683528 -0.0944569 0.175031 -0.201563 0.236234 0.301948 -0.142245 -0.350407 -0.19337 0.0473501 -0.0725593 -0.27747 0.103371 0.17572 0.192925 -0.191402 0.0326149 -0.0980565 -0.0820645 0.00794565 0.209749 0.24653 0.00570766 -0.136523 -0.247766 -0.182503 0.114115 0.057387
46 | 
47 | 12 0.494217
48 | 4 8 20 20 -0.350114 -0.023392 -0.116595 0.14089 0.350114 0.197126 -0.115137 0.0161087 0.162557 0.311529 0.295901 -0.306125 -0.280004 0.0828579 0.0240478 0.017277 0.0287732 0.135523 0.239205 -0.072818 -0.126021 -0.168983 -0.0318473 0.0748753 0.0892865 -0.0914888 0.0129376 0.210305 0.185903 0.0386759 0.166132 0.0603828 0.0816007
49 | 20 16 20 20 -0.305171 0.275763 0.250403 0.21732 0.15181 -0.00305113 0.0870038 -0.060972 -0.180267 0.159278 0.0873264 0.0175999 -0.0730725 0.0130306 0.0148139 0.0476278 -0.0794145 -0.0983501 -0.102408 0.344982 0.284974 -0.0148805 -0.0322935 0.0569539 0.141157 0.303042 0.282452 -0.114085 -0.208085 -0.228147 -0.203625 0.150856 0.086775
50 | 24 0 24 6 -0.322346 -0.255549 -0.355782 0.355782 0.220901 -0.139889 -0.0515119 0.27079 -0.105667 0.0327626 0.110244 0.355782 0.238562 0.0179007 0.00282076 0.132445 0.0341768 0.0462684 -0.0204139 -0.0245571 -0.0056198 0.261905 0.17474 -0.166816 -0.0821977 0.0586865 0.114381 0.0214783 0.0322451 -0.115689 0.107043 0.0702096 0.105872
51 | 0 8 30 20 -0.145257 0.0591194 0.10682 -0.0135948 0.00426894 -0.0217504 0.052901 -0.140989 -0.0572538 0.095878 0.171993 0.139192 0.13776 -0.200788 -0.110515 0.062712 0.223466 0.273362 0.294822 -0.15406 -0.353198 0.103017 0.247999 -0.184306 -0.353198 -0.189416 -0.135244 0.135421 0.312607 0.148005 -0.00601531 -0.10057 0.0172064
52 | 8 0 24 6 -0.347822 -0.291924 -0.256025 0.312236 0.216657 0.0604635 0.194246 -0.0834702 0.00459995 0.305514 0.133144 0.181512 0.236388 0.15208 0.170867 -0.0487316 -0.0360566 -0.0279864 -0.0635481 -0.06072 0.105931 0.07271 -0.0631185 -0.0285741 0.204921 0.161547 0.158238 0.0324198 0.0418159 -0.200216 -0.174554 0.234481 0.13524
53 | 20 20 20 20 -0.13472 0.0614852 -0.0527151 0.151573 0.0466993 0.150653 0.00133469 0.0223401 -0.0343905 0.245624 0.21159 -0.109649 -0.351236 -0.056878 0.159105 -0.114496 -0.225453 -0.18784 -0.0590177 0.148145 0.196825 -0.0785802 -0.0955774 0.0898566 0.281209 0.24167 0.251985 -0.125181 -0.24876 -0.321401 -0.183613 0.176768 0.113888
54 | 16 0 40 10 -0.361034 0.334979 0.20275 -0.0324434 -0.0853723 0.0301845 0.0421697 0.236312 -0.0458499 0.32812 0.23111 -0.250706 -0.168794 0.332085 0.117965 0.0430496 0.0838108 -0.109094 -0.0137497 0.16407 0.234564 0.0620449 0.146128 0.0493254 0.0890013 -0.101334 -0.0889906 0.211311 0.162014 0.103667 0.110429 0.0115413 0.0967582
55 | 0 4 24 12 0.0451015 0.240316 0.272003 -0.202632 -0.0392098 0.0281714 -0.174782 -0.152276 -0.0554882 -0.194126 -0.339712 0.258546 0.351487 -0.149729 0.0338817 0.0812344 -0.0312354 0.0708347 0.115788 -0.296882 -0.351487 -0.0950585 0.010032 -0.0792962 -0.0305238 0.00294667 0.00407758 0.219318 0.000834375 -0.251255 -0.124802 0.0483294 -0.0746947
56 | 12 0 12 24 -0.279076 -0.125609 -0.161619 0.200119 0.0303237 -0.0612717 -0.00509071 -0.00655398 -0.352814 -0.10269 -0.101 0.137073 0.129846 0.303076 0.0187939 0.0627639 -0.0438054 0.334997 0.282611 -0.140358 0.0441249 0.0822126 -0.0112898 -0.169411 0.0816664 0.0188293 0.0890317 0.367435 0.367435 -0.0532864 -0.0819672 -0.0309335 0.0232551
57 | 0 28 12 24 -0.189819 0.226814 0.200004 -0.147554 -0.118167 0.0991041 -0.0234189 -0.0816154 -0.0927647 -0.138793 -0.185974 0.300482 0.262065 0.033766 0.107072 0.0907754 0.11436 0.138958 0.157678 0.0142451 -0.100276 0.148171 0.357802 -0.357802 -0.357802 -0.185709 -0.0272484 0.0860602 0.0556406 0.0997465 0.0964185 -0.152961 -0.0708586
58 | 28 0 24 12 -0.0261245 -0.240536 -0.0312289 0.237712 0.206684 -0.138881 -0.0383652 0.171022 -0.06053 0.301209 0.291115 -0.203054 -0.0280476 0.0384242 -0.00430804 -0.160924 -0.145178 -0.131348 -0.0845734 0.221469 0.179261 0.165776 0.0853075 -0.265835 -0.192982 0.127937 0.0854628 -0.174423 -0.219367 -0.210165 -0.256173 0.22434 0.0335937
59 | 8 12 12 24 -0.357471 0.0226532 0.1966 -0.204673 -0.0198847 -0.17648 -0.166915 -0.0371053 0.357471 0.0912041 0.123838 -0.198569 -0.0757071 0.296168 0.0750395 -0.0232845 0.0572527 0.280497 0.188108 0.100131 0.123509 0.123491 0.0226841 0.211321 0.0587783 0.244072 0.297259 0.069033 -0.109892 -0.042628 0.224891 0.0751963 -0.00529795
60 | 
61 | 11 0.501896
62 | 16 0 40 10 -0.36408 0.15258 0.101424 0.0833005 -0.0168811 -0.0318756 0.108278 0.282904 0.0554205 0.215939 0.174879 -0.132414 -0.156742 0.0626217 0.0380281 0.141287 0.111724 -0.0430678 0.0120495 0.309095 0.297873 -0.057141 0.0687419 0.122787 0.160774 -0.204319 -0.241025 0.291239 0.313324 0.0605657 -0.0491229 0.133712 0.144125
63 | 0 4 24 12 -0.104746 0.224605 0.248393 -0.133856 -0.0209492 0.0321207 -0.0403092 -0.107036 -0.00765716 -0.138525 -0.240074 0.337817 0.378319 0.117782 0.0198304 -0.0715649 0.00718726 0.196546 0.221081 -0.298745 -0.378319 -0.0923024 0.0718807 -0.0300791 -0.111142 0.100297 0.162202 -0.0873994 -0.0497534 -0.200603 -0.207689 -0.00442005 0.0482951
64 | 16 0 12 18 -0.295388 -0.243615 -0.323047 0.176882 0.177516 -0.0861459 -0.0859016 0.00951936 -0.107609 -0.128942 -0.119096 0.231251 0.168713 0.116795 0.0481264 0.024705 -0.077268 0.305531 0.334282 -0.110468 0.0550536 0.174956 0.125566 -0.0781699 0.0436398 0.117653 0.126008 0.345192 0.289499 -0.0572135 -0.0528493 -0.0221917 0.0247499
65 | 20 20 20 20 -0.153247 0.112604 0.0481857 0.234072 0.183201 0.0162154 0.0569281 -0.0403809 -0.0960467 0.251192 0.235157 -0.182951 -0.347035 -0.0176958 0.0939955 -0.0510897 -0.179406 -0.037729 0.0231931 0.198943 0.149829 -0.113341 -0.110877 0.0904338 0.127925 0.263406 0.233778 -0.174944 -0.218249 -0.294214 -0.268934 0.220564 0.0956089
66 | 8 16 18 12 -0.369861 0.136318 0.072119 0.0385841 0.133256 -0.0399827 -0.20782 0.146931 0.173086 0.236429 0.249394 0.180126 -0.057215 0.0498329 0.252934 0.245661 -0.0470899 0.0364667 0.221979 0.0851342 0.110113 0.145054 0.198451 -0.097796 -0.0126808 -0.137273 0.0176917 0.316977 0.205807 0.189215 0.297858 -0.0313011 0.0457967
67 | 0 28 12 12 0.0680039 0.153091 0.244866 -0.253471 -0.363248 0.114377 0.0709097 -0.167637 -0.266593 0.175088 -0.169145 0.16867 -0.0356424 0.0105527 -0.0742993 -0.0674443 0.121645 0.252896 0.15706 0.0179341 -0.0370604 -0.0677263 0.0306654 -0.264124 -0.263253 0.161304 0.152278 -0.00832874 -0.0453746 0.019808 0.0433295 -0.316185 -0.291823
68 | 8 0 20 30 -0.360511 0.0200571 -0.00238614 0.209716 -0.0610175 -0.0551026 0.337889 0.267478 -0.0212667 0.0805933 0.0439167 0.175794 0.0938906 0.139751 0.113276 0.300163 0.0839932 0.192218 0.205927 -0.142312 -0.130947 -0.12014 -0.206225 0.209475 0.351974 -0.0605281 -0.129235 -0.018106 0.161138 0.174295 -0.104536 -0.0847706 0.0877826
69 | 20 8 30 20 -0.19336 0.177163 0.160357 0.122283 0.0536931 0.282664 0.227721 -0.129308 -0.259022 0.177385 0.0735432 -0.0345669 -0.0732241 -0.108036 -0.165991 0.245346 0.130737 -0.188396 -0.187826 0.223217 0.34733 -0.0590843 -0.100606 -0.0259392 0.146474 0.178043 0.219757 -0.172523 -0.298785 -0.155197 -0.00832206 0.044838 -0.00634061
70 | 8 0 12 24 -0.347245 -0.335461 -0.34073 0.222607 0.301418 0.0211971 0.114255 -0.125751 -0.011474 -0.170862 -0.110674 0.168967 0.177461 0.0216463 -0.0472234 0.17118 -0.0427571 0.134391 0.234953 0.0858471 0.236625 -0.0162319 0.0704797 -0.0665457 0.195661 0.0722552 0.173292 0.0443639 0.0478315 0.109872 -0.0936433 0.178325 0.242812
71 | 16 20 6 24 -0.147179 0.0630919 -0.0890157 0.374686 0.249938 0.0667531 -0.0338068 -0.195566 0.0592802 0.194606 0.18132 0.0137468 0.129485 -0.172793 -0.117614 0.0521887 0.124041 0.0745792 0.00678352 0.164526 0.0250539 0.0908096 0.184667 -0.254802 -0.374686 0.249393 0.227299 -0.12372 -0.374686 0.00257571 0.058177 0.0235778 -0.0256712
72 | 16 16 24 6 -0.314589 0.0439781 0.138951 -0.177242 -0.339968 0.00439803 0.22043 0.28143 0.249194 0.0497972 -0.0736346 0.15044 -0.072984 0.0464874 -0.0562933 0.0865363 0.0255786 -0.0191462 -0.112286 0.257071 0.0896218 0.154222 0.0525398 0.183625 0.27004 -0.313622 -0.296188 0.135173 0.10456 0.177458 0.125705 -0.00200204 0.0281244
73 | 
74 | 13 0.498783
75 | 4 4 30 20 -0.229091 -0.100362 -0.2005 0.130066 0.211229 0.178427 -0.0105611 0.118665 0.10109 0.295598 0.292984 -0.363277 -0.363277 0.0729639 0.154106 0.079804 -0.00453108 0.135464 0.171996 -0.17434 -0.157159 0.0252913 0.0651952 -0.00984708 0.0372692 -0.239464 -0.195049 0.192376 0.150013 0.0768058 0.140118 0.068458 0.0378611
76 | 20 16 20 20 -0.302525 0.142191 0.173323 0.240842 0.229552 0.0191278 0.140736 -0.11273 -0.221602 0.119987 0.0223615 0.0883591 -0.134353 -0.0239459 0.0726378 0.110461 -0.1396 0.133245 0.113511 0.23332 0.214139 -0.0344954 -0.0237653 0.0109188 0.0347429 0.299776 0.284661 -0.0422085 -0.157119 -0.309522 -0.248648 0.248886 0.162429
77 | 0 4 24 12 -0.0381498 0.283527 0.248006 -0.192572 -0.0377365 -0.0557831 -0.151403 -0.0905202 -0.0356861 -0.0404909 -0.20216 0.288795 0.290414 -0.0677802 0.0568397 -0.0229307 -0.0713544 0.0948624 0.146839 -0.35098 -0.33379 0.0710056 0.0429645 -0.106932 0.0257461 -0.137289 -0.0335366 0.153568 0.0425041 -0.326807 -0.138437 0.259589 0.160992
78 | 8 8 12 24 -0.378982 0.306829 0.254928 0.0304111 0.0904394 -0.0360663 -0.0581571 -0.0325614 0.313936 0.141864 0.169162 0.0509442 0.249444 0.359817 -0.028818 0.0298694 0.113568 0.0208818 0.0657478 0.0389383 0.0440995 0.0118994 -0.007708 0.24439 0.270077 0.0123836 -0.0256775 -0.054612 0.00356089 0.31011 0.26324 -0.0625163 0.00846372
79 | 0 24 12 24 -0.222395 0.0672499 0.12025 -0.0477767 -0.130475 0.198355 0.212679 -0.152851 -0.234095 -0.177315 -0.129644 0.147458 0.153709 0.151886 0.00407515 0.0222618 0.0725264 0.208887 0.197135 -0.0948499 -0.111825 0.160049 0.248679 -0.248937 -0.270979 -0.235237 -0.236685 0.239103 0.254784 0.128803 0.0086207 0.0753848 0.185758
80 | 16 0 6 24 0.0730163 0.0386967 0.0797237 0.0678749 -0.205872 -0.0744704 -0.0115581 -0.12152 -0.321905 -0.174076 -0.253687 0.124199 0.0700606 -0.133011 -0.201751 0.163209 -0.0485111 -0.262089 -0.308364 0.251792 0.198507 0.0943075 -0.0258436 0.0113415 -0.108174 0.196588 0.167019 -0.141389 -0.175448 0.209571 -0.00571164 -0.253796 -0.299698
81 | 12 12 12 24 -0.336782 0.0789678 0.112904 -0.27942 -0.245367 0.0485067 0.0522871 0.0478996 0.245285 -0.0728972 0.00328219 0.0335131 0.19419 0.0715489 0.00499987 -0.00210138 0.0838325 0.258329 0.108185 0.330446 0.314097 -0.0276721 -0.0954354 0.143433 0.0553743 0.347122 0.315476 -0.0672877 -0.121526 0.0355355 0.114856 -0.109308 -0.0998654
82 | 20 20 20 20 -0.0662569 0.0283931 -0.137111 0.15169 0.0875735 -0.0607412 -0.0305409 0.160088 0.0212801 0.19173 0.144318 -0.134025 -0.351618 -0.0357132 0.131315 -0.131979 -0.183278 -0.0671661 0.0695825 0.0731253 0.0711527 -0.165081 -0.0929893 0.0618314 0.212811 0.273312 0.267798 -0.158973 -0.336936 -0.351618 -0.195013 0.272278 0.0996006
83 | 4 4 12 24 -0.356675 0.019497 0.117836 0.286384 0.275671 0.0534012 0.170231 -0.00710427 -0.0718629 0.267715 0.322308 -0.0293407 -0.0565582 -0.031858 0.0498118 0.112767 0.0183026 0.0229094 -0.106487 0.141465 0.277437 0.180264 -0.129345 -0.136349 0.102999 0.293138 0.243207 -0.105962 0.0243751 0.250964 0.0918212 -0.198436 -0.042904
84 | 0 0 6 24 -0.0928533 0.19549 0.130855 -0.172327 -0.30826 -0.268513 -0.124287 0.0528794 0.351067 -0.127708 -0.252429 0.24992 0.106267 0.0352341 0.0757421 -0.00812923 0.139801 -0.317259 -0.0954179 0.264929 0.178775 0.0178959 0.00659988 -0.138013 -0.159935 -0.0250608 -0.178045 0.195575 0.192797 0.133338 -0.0981471 -0.0189815 0.12154
85 | 0 16 24 6 0.0385782 -0.133513 -0.154937 0.16485 0.115046 -0.011692 0.235721 -0.103248 -0.108457 0.21825 0.145522 -0.0611039 -0.357613 -0.0131616 -0.132178 -0.0430153 -0.0934175 0.120632 0.0225865 -0.155457 -0.319005 0.134062 0.0963147 -0.0912014 -0.0953387 0.292065 0.274691 -0.0601385 -0.082206 -0.0576181 0.197094 -0.26853 -0.357613
86 | 0 8 12 24 -0.278565 0.146123 0.137407 0.304141 0.347757 0.121629 0.180915 -0.136458 -0.293986 0.323808 0.332708 -0.08198 0.0579564 0.0282133 -0.186334 0.0766941 -0.00227364 -0.0571502 0.00137349 -0.18906 -0.000387721 0.0529434 0.00536383 -0.0204336 0.156085 0.0463233 0.116395 -0.00960852 -0.158167 -0.179148 0.0603636 0.23662 0.200001
87 | 32 16 24 6 0.0577473 -0.298664 -0.281051 0.266062 0.0455574 0.00035151 -0.0135557 0.301201 -0.028636 0.0596 0.0848259 -0.032205 -0.258337 0.0618552 0.00312378 -0.121183 -0.184822 0.0357844 0.270088 -0.0531704 -0.164134 -0.016061 0.0503472 -0.129345 -0.0667124 0.191182 0.180473 -0.0590233 -0.071379 -0.360236 -0.360236 0.221879 0.0868195


--------------------------------------------------------------------------------
/data/outColor1.mp4:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/data/outColor1.mp4


--------------------------------------------------------------------------------
/data/outColor2.mp4:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/data/outColor2.mp4


--------------------------------------------------------------------------------
/data/resultMatrix.txt:
--------------------------------------------------------------------------------
1 | 25 3 0 0 1 1 
2 | 2 27 0 0 0 0 
3 | 0 0 30 1 0 1 
4 | 0 0 1 28 2 0 
5 | 1 0 0 0 27 3 
6 | 0 0 0 1 0 29 
7 | 


--------------------------------------------------------------------------------
/data/zInput12/theanocnn.json:
--------------------------------------------------------------------------------
1 | [[[[0.1579788173043373, 0.2961204466666364, 0.28052473452509413, 0.5822940516366423, 0.23810391721900345, -0.18805271798528092, 0.3146650823401803, 0.5014556627270617, -0.4721984747535301, -0.013835434495267014, -0.10912522787862454, 0.03544907182826434, -0.5292658229872349, -1.0394350028967787, -1.0908152507138793, -0.9913887610119967, 1.5013740304639245, -0.3740580600739958, -0.20015490029215668, 0.11883279079213156, -0.11218523119003894, 0.5817701691204245, 0.4050745862166843, 0.05025088716044678, 0.5136682626537493], [0.6688310522561846, 0.146718806265588, -0.3647473281388782, -0.38416767222913484, 0.25676375900993403, 1.1697592423125427, -0.17037731832706143, 0.1473627506205251, -0.07370328114220753, 0.8539706192860157, 1.1519737725860706, 0.4092907675386512, 0.5262466489965649, 0.20702942768787955, 0.6891512439380896, 0.35389094127487325, 0.5427539111954495, -0.6122955943988614, 0.6000489485017794, 0.5771266707088482, 0.40331926552535186, -0.2816343684364563, 0.4820877852488716, -0.29316183489736625, 0.16199586509359176]], [-0.4370692627954934, -2.0589014322643617]], [[[-0.48963591314475763, -1.1704027996297208, -0.00875929118851474, 0.06653122925810545, -0.4933454937964717, -0.8886569381250906, -0.43581307312592854, -0.5369922214935475], [0.48963591314475774, 1.1704027996297217, 0.00875929118851474, -0.0665312292581056, 0.4933454937964713, 0.8886569381250897, 0.4358130731259282, 0.5369922214935475]], [0.7342387374322045, -0.7342387374322038]]]


--------------------------------------------------------------------------------
/data/zInput20/theanocnn.json:
--------------------------------------------------------------------------------
1 | [[[[1.942323227521379, 1.6344067369856827, 1.0141819807392078, 1.1310550585873418, 1.7117012610542, -0.24284751642910338, -0.30949706347125067, -0.2283990427289676, -0.8662084116541944, 0.7789345885255776, 0.18681186388614504, -0.44927674627342823, -0.018752105786115837, -1.0972116100847242, -0.8904160235863825, -2.7809806397476504, -2.46484462708447, -2.8083734377120217, -1.9241069184842396, 0.1505409072775627, 2.2112556556393526, 1.5063184093169835, 1.753687886140495, 0.34689453775739537, 1.2550582619897206], [1.625037636203617, -0.5714643626614894, -1.8687694029985311, -3.939469013233547, -6.209397364979733, 0.2467112818589018, -1.0615777394045325, 1.47772568024536, 4.110198415536089, 4.55605638564367, 1.0623097155303738, 0.7356993391200247, -0.4545336237127385, -0.8391349993009636, -1.0792827210230884, 0.6685605960281255, 0.8119193763857344, -0.21064725355065267, 0.27762498455414447, 0.6519431868304306, -0.12642772064833613, 0.14454681730138916, 0.8335164296476318, 0.1600515193886391, 1.0256430618655539]], [-1.937708014038304, -1.5388614698081702]], [[[-0.056196787653649494, -0.5906331102852002, -0.18975034469347973, -0.28592784940879273, 0.6492228508791698, 0.1676898836347293, 0.4479400592452021, 0.5022019425416557, 0.47468175794084494, -0.39530046600456026, -0.5888043043740578, 0.32961196105450546, 0.48267764475452, -0.3760410929044946, -0.5549173719672085, 0.25485971631979026, -0.3403232219374212, 0.44286014681065455, -0.28482691008183564, -0.18758999563446505, -0.8329800602162615, 0.2603172053510445, -0.7341312070789388, -0.47356497559792804, 0.23449573486428626, -0.0016274137564284109, -0.09753301052591332, 0.40784059932118855, 0.0007526822697131558, -0.6282986650419908, -0.37990191210993185, 0.37047692405784066], [0.0561967876536496, 0.5906331102852003, 0.1897503446934797, 0.2859278494087931, -0.6492228508791696, -0.16768988363472956, -0.44794005924520214, -0.5022019425416561, -0.4746817579408449, 0.3953004660045603, 0.5888043043740573, -0.3296119610545055, -0.48267764475452, 0.37604109290449467, 0.5549173719672087, -0.2548597163197901, 0.34032322193742137, -0.4428601468106544, 0.2848269100818351, 0.18758999563446474, 0.8329800602162611, -0.2603172053510441, 0.7341312070789383, 0.4735649755979278, -0.23449573486428654, 0.001627413756428382, 0.09753301052591337, -0.40784059932118866, -0.0007526822697131347, 0.6282986650419906, 0.3799019121099316, -0.370476924057841]], [2.294155429214866, -2.2941554292148623]]]


--------------------------------------------------------------------------------
/data/zInput24/theanocnn.json:
--------------------------------------------------------------------------------
1 | [[[[-0.0013492353702092691, -0.30225785163351326, -0.6032477836556043, -0.3280587922546115, 0.25500504767135657, 0.9629545154309812, 1.4624325173020751, 1.7486705987617153, 1.0111238321030511, 1.1970083427132114, 1.3318436382398564, 0.748358897446316, 1.0083073859033738, 0.3230122854465798, -0.6230591427374722, -2.3042441552825057, -2.872763545562979, -3.9455794680831957, -2.589817013395963, -0.592026411714205, 0.8457634536311359, 0.8721067548917225, 1.7009193199403114, 0.7108804789196536, 0.7119733253292257], [1.0294777031672278, 1.701576726321337, 0.7643690296731365, -2.1989445333163022, -3.5572832198864006, 0.5532241426029691, -0.0010424505647128534, 0.19060944430319227, -0.841699761485703, 0.20173602547607758, 0.7796996744060596, 0.6128033625764033, 0.3393717532490747, -0.4018431152326931, -0.1283143112004536, 0.3104061937195179, 0.9963445356302851, -0.03897233384464926, 0.08848370169558618, -0.342548241197508, 0.5843732631181454, 0.5538230277537854, 0.6900521326356991, -0.2998931590273716, 0.012695947095889447]], [-1.2431093962512167, -1.1786487819582263]], [[[-0.15284699806294394, -0.5703539206866449, 0.181771558035712, -0.3643463405300111, -0.312927201800944, 0.18650435649988192, 0.022557116826057183, 0.7256111468827519, 0.24723225285193298, 0.10913310750978863, 0.4845521734901893, -0.054238091631719304, -0.5054541743292827, -0.20115802447134304, 0.46283711661448745, 0.4052154816717093, -0.09542783781816946, -0.6417470607447102, -0.3599066861824806, 0.45483597349699434, 0.519527655439504, -0.06195835392587051, -0.5284512252899455, -0.30198724768363006, 0.352972966038185, 0.2597023658397933, 0.6935912425564739, 0.1040980081085055, -0.19487651801869488, 0.10563653998217504, -0.6478143134777876, -0.46733346148513694, -0.6022883921105181, -0.7335151562436539, -0.10607961708011666, 0.10821683185429117, -0.2277594679072588, -0.029831928748730478, 0.12412244593493442, -0.1428821510211233, 0.08516659838453487, -0.07821443035064346, 0.1622401909067941, 0.4327308787968827, -0.0034441474469255073, -0.1179258480655713, -0.1610904354248086, -0.06814044593080683, -0.15849762179478222, -0.19595700157137533], [0.15284699806294405, 0.5703539206866449, -0.18177155803571204, 0.36434634053001114, 0.3129272018009438, -0.18650435649988187, -0.022557116826057152, -0.7256111468827516, -0.24723225285193304, -0.10913310750978863, -0.4845521734901892, 0.0542380916317192, 0.5054541743292824, 0.20115802447134293, -0.46283711661448734, -0.4052154816717091, 0.09542783781816946, 0.6417470607447088, 0.35990668618248073, -0.45483597349699423, -0.5195276554395043, 0.06195835392587051, 0.5284512252899451, 0.30198724768362994, -0.3529729660381852, -0.25970236583979317, -0.6935912425564739, -0.10409800810850552, 0.194876518018695, -0.10563653998217522, 0.6478143134777875, 0.46733346148513644, 0.6022883921105194, 0.7335151562436529, 0.10607961708011676, -0.10821683185429101, 0.2277594679072587, 0.029831928748730346, -0.12412244593493428, 0.14288215102112314, -0.08516659838453496, 0.0782144303506436, -0.16224019090679417, -0.43273087879688266, 0.003444147446925484, 0.11792584806557117, 0.16109043542480866, 0.06814044593080675, 0.15849762179478208, 0.19595700157137508]], [2.1889834246797166, -2.1889834246797144]]]


--------------------------------------------------------------------------------
/data/zInput40/.DS_Store:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/data/zInput40/.DS_Store


--------------------------------------------------------------------------------
/data/zInput40/参数1/.DS_Store:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/data/zInput40/参数1/.DS_Store


--------------------------------------------------------------------------------
/data/zInput40/参数2/.DS_Store:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/data/zInput40/参数2/.DS_Store


--------------------------------------------------------------------------------
/data/zInput40/参数3/.DS_Store:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/data/zInput40/参数3/.DS_Store


--------------------------------------------------------------------------------
/featuremap.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/featuremap.cpp


--------------------------------------------------------------------------------
/featuremap.h:
--------------------------------------------------------------------------------
 1 | #ifndef FEATUREMAP_H  
 2 | #define FEATUREMAP_H  
 3 |   
 4 | #include <cassert>  
 5 | #include <vector>  
 6 | using std::vector;  
 7 |   
 8 | class FeatureMap  
 9 | {  
10 | public:  
11 |     FeatureMap(int iInputImageNumber, int iInputImageWidth, int iFeatureMapNumber, int iKernelWidth);  
12 |     ~FeatureMap();  
13 |     void Forward_propagation(double* );  
14 |     void Back_propagation(double* , double* , double );   
15 |   
16 |   
17 |     void Convolute(double *pdInputData);  
18 |   
19 |     int GetFeatureMapSize()  
20 |     {  
21 |         return m_iFeatureMapSize;  
22 |     }  
23 |   
24 |     int GetFeatureMapWidth()  
25 |     {  
26 |         return m_iFeatureMapWidth;  
27 |     }  
28 |   
29 |     double* GetFeatureMapValue()  
30 |     {  
31 |         assert(m_pdOutputValue != NULL);  
32 |         return m_pdOutputValue;  
33 |     }  
34 |   
35 |     void SetWeigh(const vector<double *> &vecWeigh);  
36 |     void PrintOutputData();  
37 |   
38 |     double **m_ppdWeigh;  
39 |     double *m_pdBias;  
40 |   
41 | private:  
42 |     int m_iInputImageNumber;  
43 |     int m_iInputImageWidth;  
44 |     int m_iInputImageSize;  
45 |     int m_iFeatureMapNumber;  
46 |     int m_iFeatureMapWidth;  
47 |     int m_iFeatureMapSize;  
48 |     int m_iKernelWidth;  
49 |       
50 | //    double m_dBias;  
51 |     double *m_pdOutputValue;  
52 | };  
53 |   
54 | #endif // FEATUREMAP_H  
55 | 


--------------------------------------------------------------------------------
/head/.DS_Store:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/.DS_Store


--------------------------------------------------------------------------------
/head/libopencv_calib3d.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_calib3d.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_calib3d.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_calib3d.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_calib3d.dylib:
--------------------------------------------------------------------------------
1 | libopencv_calib3d.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_contrib.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_contrib.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_contrib.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_contrib.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_contrib.dylib:
--------------------------------------------------------------------------------
1 | libopencv_contrib.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_core.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_core.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_core.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_core.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_core.dylib:
--------------------------------------------------------------------------------
1 | libopencv_core.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_features2d.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_features2d.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_features2d.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_features2d.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_features2d.dylib:
--------------------------------------------------------------------------------
1 | libopencv_features2d.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_flann.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_flann.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_flann.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_flann.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_flann.dylib:
--------------------------------------------------------------------------------
1 | libopencv_flann.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_gpu.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_gpu.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_gpu.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_gpu.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_gpu.dylib:
--------------------------------------------------------------------------------
1 | libopencv_gpu.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_highgui.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_highgui.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_highgui.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_highgui.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_highgui.dylib:
--------------------------------------------------------------------------------
1 | libopencv_highgui.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_imgproc.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_imgproc.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_imgproc.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_imgproc.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_imgproc.dylib:
--------------------------------------------------------------------------------
1 | libopencv_imgproc.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_legacy.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_legacy.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_legacy.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_legacy.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_legacy.dylib:
--------------------------------------------------------------------------------
1 | libopencv_legacy.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_ml.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_ml.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_ml.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_ml.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_ml.dylib:
--------------------------------------------------------------------------------
1 | libopencv_ml.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_nonfree.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_nonfree.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_nonfree.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_nonfree.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_nonfree.dylib:
--------------------------------------------------------------------------------
1 | libopencv_nonfree.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_objdetect.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_objdetect.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_objdetect.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_objdetect.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_objdetect.dylib:
--------------------------------------------------------------------------------
1 | libopencv_objdetect.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_ocl.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_ocl.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_ocl.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_ocl.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_ocl.dylib:
--------------------------------------------------------------------------------
1 | libopencv_ocl.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_photo.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_photo.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_photo.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_photo.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_photo.dylib:
--------------------------------------------------------------------------------
1 | libopencv_photo.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_stitching.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_stitching.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_stitching.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_stitching.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_stitching.dylib:
--------------------------------------------------------------------------------
1 | libopencv_stitching.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_superres.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_superres.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_superres.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_superres.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_superres.dylib:
--------------------------------------------------------------------------------
1 | libopencv_superres.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_ts.a:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_ts.a


--------------------------------------------------------------------------------
/head/libopencv_video.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_video.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_video.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_video.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_video.dylib:
--------------------------------------------------------------------------------
1 | libopencv_video.2.4.dylib


--------------------------------------------------------------------------------
/head/libopencv_videostab.2.4.13.dylib:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/libopencv_videostab.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_videostab.2.4.dylib:
--------------------------------------------------------------------------------
1 | libopencv_videostab.2.4.13.dylib


--------------------------------------------------------------------------------
/head/libopencv_videostab.dylib:
--------------------------------------------------------------------------------
1 | libopencv_videostab.2.4.dylib


--------------------------------------------------------------------------------
/head/pkgconfig/opencv.pc:
--------------------------------------------------------------------------------
 1 | # Package Information for pkg-config
 2 | 
 3 | prefix=/usr/local/Cellar/opencv/2.4.13.1
 4 | exec_prefix=${prefix}
 5 | libdir=${exec_prefix}/lib
 6 | includedir_old=${prefix}/include/opencv
 7 | includedir_new=${prefix}/include
 8 | 
 9 | Name: OpenCV
10 | Description: Open Source Computer Vision Library
11 | Version: 2.4.13.1
12 | Libs: -L${exec_prefix}/lib -lopencv_calib3d -lopencv_contrib -lopencv_core -lopencv_features2d -lopencv_flann -lopencv_gpu -lopencv_highgui -lopencv_imgproc -lopencv_legacy -lopencv_ml -lopencv_nonfree -lopencv_objdetect -lopencv_ocl -lopencv_photo -lopencv_stitching -lopencv_superres -lopencv_ts -lopencv_video -lopencv_videostab
13 | Cflags: -I${includedir_old} -I${includedir_new}
14 | 


--------------------------------------------------------------------------------
/head/python2.7/site-packages/cv.py:
--------------------------------------------------------------------------------
1 | from cv2.cv import *
2 | 


--------------------------------------------------------------------------------
/head/python2.7/site-packages/cv2.so:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/head/python2.7/site-packages/cv2.so


--------------------------------------------------------------------------------
/hiddenLayer.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/hiddenLayer.cpp


--------------------------------------------------------------------------------
/hiddenLayer.h:
--------------------------------------------------------------------------------
 1 | #ifndef HIDDENLAYER_H  
 2 | #define HIDDENLAYER_H  
 3 |    
 4 | #include "neuralbase.h"  
 5 |   
 6 | class HiddenLayer: public NeuralBase  
 7 | {  
 8 | public:  
 9 |     HiddenLayer(int n_i, int n_o);  
10 |     ~HiddenLayer();  
11 |   
12 |     double* Forward_propagation(double* input_data);  
13 |       
14 |     void Back_propagation(double *pdInputData, double *pdNextLayerDelta,  
15 |                           double** ppdnextLayerW, int iNextLayerOutNum, double dLr);  
16 |       
17 |   
18 |   
19 | };  
20 |   
21 | #endif  


--------------------------------------------------------------------------------
/logisticRegression.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/logisticRegression.cpp


--------------------------------------------------------------------------------
/logisticRegression.h:
--------------------------------------------------------------------------------
 1 | 
 2 | #ifndef LOGISTICREGRESSIONLAYER  
 3 | #define LOGISTICREGRESSIONLAYER  
 4 | #include "neuralbase.h"  
 5 |   
 6 | typedef unsigned short WORD;  
 7 |   
 8 | class LogisticRegression: public NeuralBase  
 9 | {   
10 | public:  
11 |     LogisticRegression(int n_i, int i_o, int);  
12 |     ~LogisticRegression();  
13 |   
14 |     double* Forward_propagation(double* input_data);  
15 |     void Softmax(double* x);  
16 |     void Train(double *pdTrain, WORD usLabel, double dLr);  
17 |   
18 |   
19 |     void SetOldWb(double ppdWeigh[][3], double arriBias[8]);  
20 |     int Predict(double *);  
21 |   
22 |     void MakeLabels(int* piMax, double (*pplabels)[8]);  
23 | };  
24 |   
25 | void Test_lr();  
26 | void Testwb();  
27 | void Test_theano(const int m_iInput, const int m_iOut);  
28 | #endif  
29 | 


--------------------------------------------------------------------------------
/main.cpp:
--------------------------------------------------------------------------------
  1 | //
  2 | //  main.cpp
  3 | //  faceDetection-cnn
  4 | //
  5 | //  Created by 刘鹏 on 2016/11/27.
  6 | //  Copyright © 2016年 刘鹏. All rights reserved.
  7 | //
  8 | #include <iostream>
  9 | #include <opencv2/core/core.hpp>
 10 | #include <opencv2/highgui/highgui.hpp>
 11 | #include <opencv2/opencv.hpp>
 12 | // 新版本写在下面文件中：
 13 | #include <opencv2/nonfree/features2d.hpp>
 14 | //#include "opencv2/features2d/features2d.hpp"
 15 | #include<opencv2/legacy/legacy.hpp>
 16 | 
 17 | #include <iostream>
 18 | #include <stdio.h>
 19 | 
 20 | #include "FDImage.h"
 21 | #include "CCCascadeClassifier.h"
 22 | #include <time.h>
 23 | #include <stdio.h>
 24 | #include <algorithm>
 25 | 
 26 | using namespace std;
 27 | using namespace cv;
 28 | 
 29 | 
 30 | IplImage* cutImage(IplImage* src, CvRect rect) {
 31 |     cvSetImageROI(src, rect);
 32 |     IplImage* dst = cvCreateImage(cvSize(rect.width, rect.height),
 33 |                                   src->depth,
 34 |                                   src->nChannels);
 35 |     
 36 |     cvCopy(src,dst,0);
 37 |     cvResetImageROI(src);
 38 |     // imshow("dst",(Mat)dst);
 39 |     return dst;
 40 | }
 41 | 
 42 | // Function main
 43 | int main()
 44 | {
 45 |     
 46 |     FDImage* image;
 47 |     image = new FDImage;
 48 |     vector<Rect> vFaceRect;
 49 |     vector<Rect> faceBox;
 50 |     CCCascadeClassifier *cascade;
 51 |     cascade = new CCCascadeClassifier;
 52 |     //========================================================
 53 |     // CvCapture  «“ª∏ˆΩ·ππ㨔√¿¥±£¥ÊÕºœÒ≤∂ªÒÀ˘–Ë“™µƒ–≈œ¢°£
 54 |     // opencv÷π©¡Ω÷÷∑Ω Ω¥”Õ‚≤ø≤∂ªÒÕºœÒ£¨“ª÷÷ «¥”…„œÒÕ∑÷–£¨“ª÷÷
 55 |     //  «Õ®π˝Ω‚¬Î ”∆µµ√µΩÕºœÒ°£¡Ω÷÷∑Ω Ω∂º±ÿ–Î¥”µ⁄“ª÷°ø™ º“ª÷°“ª÷°
 56 |     // µƒ∞¥À≥–ÚªÒ»°£¨“Ú¥À√øªÒ»°“ª÷°∫Û∂º“™±£¥Êœ‡”¶µƒ◊¥Ã¨∫Õ≤Œ ˝°£
 57 |     // ±»»Á¥” ”∆µŒƒº˛÷–ªÒ»°£¨–Ë“™±£¥Ê ”∆µŒƒº˛µƒŒƒº˛√˚£¨œ‡”¶µƒ******
 58 |     // ¿‡–Õ£¨œ¬“ª¥Œ»Áπ˚“™ªÒ»°Ω´–Ë“™Ω‚¬Îƒƒ“ª÷°µ»°£ ’‚–©–≈œ¢∂º±£¥Ê‘⁄
 59 |     // CvCaptureΩ·ππ÷–£¨√øªÒ»°“ª÷°∫Û£¨’‚–©–≈œ¢∂ºΩ´±ª∏¸–¬£¨ªÒ»°œ¬“ª÷°
 60 |     // –Ë“™Ω´–¬–≈œ¢¥´∏¯ªÒ»°µƒapiΩ”ø⁄
 61 |     //=======================================================
 62 |     CvCapture* capture = 0;
 63 |     //===========================================================
 64 |     // IplImage  «Ω·ππÿ‡–Õ£¨”√¿¥±£¥Ê“ª÷°ÕºœÒµƒ–≈œ¢£¨“≤æÕ «“ª÷°
 65 |     // ÕºœÒµƒÀ˘”–œÒÀÿ÷µππ≥…µƒ“ª∏ˆæÿ’Û
 66 |     //===========================================================
 67 |     IplImage *frame, *frame_copy = 0;
 68 |     IplImage *src;
 69 |     IplImage *pGrayImg;
 70 |     
 71 |     // ¥¥Ω®“ª∏ˆ¥∞ø⁄£¨”√°∞result°±◊˜Œ™¥∞ø⁄µƒ±Í ∂∑˚
 72 |     cvNamedWindow( "result", 1 );
 73 |     
 74 |     // ==========================================
 75 |     // ≥ı ºªØ“ª∏ˆ ”∆µ≤∂ªÒ≤Ÿ◊˜°£
 76 |     // ∏ÊÀflµ◊≤„µƒ≤∂ªÒapiŒ“œÎ¥” Capture1.avi÷–≤∂ªÒÕº∆¨£¨
 77 |     // µ◊≤„apiΩ´ºÏ≤‚≤¢—°‘Òœ‡”¶µƒ******≤¢◊ˆ∫√◊º±∏π§◊˜
 78 |     //==============================================
 79 |     capture = cvCaptureFromFile("/Users/liupeng/Desktop/my/faceDetection-cnn/faceDetection-cnn/data/outColor2.mp4");
 80 |     // ªÒ»°…„œÒÕ∑
 81 |     // capture = cvCaptureFromCAM(0);
 82 |     
 83 |     // »Áπ˚ ≥ı ºªØ ß∞‹£¨ƒ«√¥captureŒ™ø’÷∏’Σ¨≥ÖÚÕ£÷π£¨
 84 |     // ∑Ò‘ÚΩ¯»Î≤∂ªÒ—≠ª∑
 85 |     if( capture )
 86 |     {
 87 |         // ≤∂ªÒ—≠ª∑
 88 |         while(1)
 89 |         {
 90 |             // µ˜”√cvGrabFrame,»√µ◊≤„apiΩ‚¬Î“ª÷°ÕºœÒ
 91 |             // »Áπ˚Ω‚¬Î ß∞‹£¨æÕÕÀ≥ˆ—≠ª∑
 92 |             // »Áπ˚≥…𶣨ق¬ÎµƒÕºœÒ±£¥Ê‘⁄µ◊≤„apiµƒª∫¥Ê÷–
 93 |             if( !cvGrabFrame( capture ))
 94 |                 break;
 95 |             
 96 |             // Ω´Ω‚¬Îµ√µΩÕºœÒ–≈œ¢¥”ª∫¥Ê÷–◊™ªª≥…IplImage∏Ò Ω∑≈‘⁄frame÷–
 97 |             // frame = cvRetrieveFrame( capture );
 98 |             frame = cvQueryFrame(capture);
 99 |             
100 |             // »Áπ˚ªÒ»°ª∫¥ÊªÚ◊™ªª ß∞‹£¨‘ÚÕÀ≥ˆ—≠ª∑
101 |             if( !frame )
102 |                 break;
103 |             
104 |             Mat mFrame = (Mat)frame;
105 |             // Apply the classifier to the frame
106 |             if (!mFrame.empty()){
107 |                 
108 |                 src = frame;
109 |                 cvShowImage( "image", src );
110 |                 
111 |                 CvRect rect;
112 |                 int iX = 900;
113 |                 int iY = 200;
114 |                 
115 |                 rect.x = 200;
116 |                 rect.y = 0;
117 |                 rect.width = 200;
118 |                 rect.height = 200;
119 |                 
120 |                 /* Infrared ÕºœÒœ¬µƒ…Ë÷√
121 |                  rect.x = 12;
122 |                  rect.y = 24;
123 |                  rect.width = 400;
124 |                  rect.height = 400; */
125 |                 
126 |                 IplImage *temp = cutImage(src,rect);
127 |                 // imshow("aaaa",Mat(temp));
128 |                 
129 |                 // µ˜’˚Õº∆¨µƒ¥Û–°°£
130 |                 IplImage *imageresize;
131 |                 imageresize=cvCreateImage(cvSize(200,200),temp->depth,temp->nChannels);
132 |                 cvResize(temp,imageresize,CV_INTER_LINEAR);
133 |                 // imshow("img1", Mat(temp));
134 |                 // waitKey();
135 |                 
136 |                 pGrayImg=cvCreateImage(cvGetSize(imageresize),8,1);
137 |                 cvCvtColor(imageresize,pGrayImg,CV_RGB2GRAY);
138 |                 
139 |                 imshow("img2",Mat(pGrayImg));
140 |                 // waitKey();
141 |                 
142 |                 // Ω´IplImage∏Ò ΩµƒÕº∆¨◊™ªØŒ™FDImage∏Ò ΩµƒÕº∆¨
143 |                 image->Load(pGrayImg);
144 |                 double t1, t2;
145 |                 //cascade->LoadDefaultCascade();
146 |                 //if(cascade ==NULL)
147 |                 ///	return;
148 |                 FDImage &im = *image;
149 |                 //if(cascade->count>0)
150 |                 {
151 |                     t1 = getTickCount();
152 |                     //cascade->FaceDetect_ScaleImage(im);
153 |                     vFaceRect = cascade->FaceDetectWithRet_ScaleIamge(im,"E:/");
154 |                     for (int ic = 0; ic < vFaceRect.size(); ic++) // Iterate through all current elements (detected faces)
155 |                     {
156 |                         if(vFaceRect[ic].x > 0)
157 |                         {
158 |                             float alpha = 1.0;
159 |                             Point pt1(200+vFaceRect[ic].x*alpha, vFaceRect[ic].y*alpha); // Display detected faces on main window - live stream from camera
160 |                             Point pt2((200+vFaceRect[ic].x+vFaceRect[ic].width)*alpha, (vFaceRect[ic].y+vFaceRect[ic].height)*alpha);
161 |                             Mat mF = (Mat)frame;
162 |                             rectangle(mF, pt1, pt2, Scalar(0, 255, 0), 2, 8, 0);
163 |                         }
164 |                     }
165 |                 }
166 |             }
167 |             else{
168 |                 printf(" --(!) No captured frame -- Break!");
169 |                 //break;
170 |             }
171 |             
172 |             int c = waitKey(1);
173 |             
174 |             if (27 == char(c)){
175 |                 // break;
176 |             }
177 |             
178 |             // Ω´frame÷–µƒÕºœÒ–≈œ¢‘⁄¥∞ø⁄result÷–œ‘ æ
179 |             cvShowImage( "gray", pGrayImg );
180 |             cvShowImage( "result", frame );
181 |             
182 |             // ‘›Õ£“ªª·∂˘£¨»√ƒ„ø¥“ªœ¬ÕºœÒ
183 |             cvWaitKey(1);
184 |             
185 |         }
186 |         
187 |         // ÕÀ≥ˆ≥ÖÚ÷Æ«∞“™«Â¿Ì“ªœ¬∂—’ª÷–µƒƒ⁄¥Ê£¨√‚µ√ƒ⁄¥Ê–π¬∂
188 |         //cvReleaseImage( &frame );◊¢“‚≤ª–Ë“™’‚扣¨“ÚŒ™frame «¥” ”∆µ÷–≤∂ªÒµƒ£¨√ª”–µ•∂¿∑÷≈‰ƒ⁄¥Ê£¨Œfi–Ë Õ∑≈£¨µ±capture  Õ∑≈µƒ ±∫Úframe◊‘»ªæÕ Õ∑≈¡À°£
189 |         
190 |         // ÕÀ≥ˆ÷Æ«∞Ω· ¯µ◊≤„apiµƒ≤∂ªÒ≤Ÿ◊˜£¨√‚µ√À¸√«’º◊≈√©ø”≤ª¿≠ ∫
191 |         // ±»»Áª· πµ√±µƒ≥Öڌfi∑®∑√Œ “—æ≠±ªÀ¸√«¥Úø™µƒŒƒº˛
192 |         cvReleaseCapture( &capture );
193 |         
194 |     }
195 |     cvDestroyWindow("result");
196 |     
197 |     return 0;
198 | }
199 | 
200 | 
201 | 
202 | 


--------------------------------------------------------------------------------
/mlp.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/mlp.cpp


--------------------------------------------------------------------------------
/mlp.h:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/mlp.h


--------------------------------------------------------------------------------
/neuralNetwork.h:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/neuralNetwork.h


--------------------------------------------------------------------------------
/neuralbase.cpp:
--------------------------------------------------------------------------------
  1 | #include "stdafx.h"
  2 | #include "neuralbase.h"  
  3 |   
  4 | #include <cmath>  
  5 | #include <cassert>  
  6 | #include <ctime>  
  7 |    
  8 | #include <iomanip>  
  9 | #include <iostream>  
 10 |   
 11 |   
 12 | #include "util.h"  
 13 |   
 14 | using namespace std;  
 15 |   
 16 |   
 17 | NeuralBase::NeuralBase(int n_i, int n_o, int n_t):m_iInput(n_i), m_iOut(n_o), m_iSamplenum(n_t)  
 18 | {  
 19 |     m_ppdW = new double* [m_iOut];  
 20 |     for(int i = 0; i < m_iOut; ++i)  
 21 |     {  
 22 |         m_ppdW[i] = new double [m_iInput];  
 23 |     }  
 24 |     m_pdBias = new double [m_iOut];  
 25 |   
 26 |     double a = 1.0 / m_iInput;  
 27 |   
 28 |     srand((unsigned)time(NULL));  
 29 |     for(int i = 0; i < m_iOut; ++i)  
 30 |     {  
 31 |         for(int j = 0; j < m_iInput; ++j)  
 32 |             m_ppdW[i][j] = uniform(-a, a);  
 33 |         m_pdBias[i] = uniform(-a, a);  
 34 |     }  
 35 |   
 36 |     m_pdDelta = new double [m_iOut];  
 37 |     m_pdOutdata = new double [m_iOut];  
 38 | }  
 39 |   
 40 | NeuralBase::~NeuralBase()  
 41 | {  
 42 |     Callbackwb();  
 43 |     delete[] m_pdOutdata;  
 44 |     delete[] m_pdDelta;  
 45 | }  
 46 | void NeuralBase::Callbackwb()  
 47 | {  
 48 |     _callbackwb();  
 49 | }  
 50 | double NeuralBase::CalErrorRate(const vector<double *> &vecvalid, const vector<WORD> &vecValidlabel)  
 51 | {  
 52 |     int iErrorNumber = 0, iValidNumber = vecValidlabel.size();  
 53 |     for (int i = 0; i < iValidNumber; ++i)  
 54 |     {  
 55 |         int iResult = Predict(vecvalid[i]);  
 56 |         if (iResult != vecValidlabel[i])  
 57 |         {  
 58 |             ++iErrorNumber;  
 59 |         }  
 60 |     }  
 61 |   
 62 |     cout << "the num of error is " << iErrorNumber << endl;  
 63 |     double dErrorRate = (double)iErrorNumber / iValidNumber;  
 64 |     cout << "the error rate of Train sample by softmax is " << setprecision(10) << dErrorRate * 100 << "%" << endl;  
 65 |   
 66 |     return dErrorRate;  
 67 | }  
 68 | int NeuralBase::Predict(double *)  
 69 | {  
 70 |     cout << "NeuralBase::Predict(double *)" << endl;  
 71 |     return 0;  
 72 | }  
 73 |   
 74 | void NeuralBase::_callbackwb()  
 75 | {  
 76 |     for(int i=0; i < m_iOut; i++)  
 77 |         delete []m_ppdW[i];  
 78 |     delete[] m_ppdW;  
 79 |     delete[] m_pdBias;  
 80 | }  
 81 |   
 82 | void NeuralBase::Printwb()  
 83 | {  
 84 |     cout << "'****m_ppdW****\n";  
 85 |     for(int i = 0; i < m_iOut; ++i)  
 86 |     {  
 87 |         for(int j = 0; j < m_iInput; ++j)  
 88 |             cout << m_ppdW[i][j] << ' ';  
 89 |         cout << endl;  
 90 |     }  
 91 |     cout << "'****m_pdBias****\n";  
 92 |     for(int i = 0; i < m_iOut; ++i)  
 93 |     {  
 94 |         cout << m_pdBias[i] << ' ';  
 95 |     }  
 96 |     cout << endl;  
 97 |     cout << "'****output****\n";  
 98 |     for(int i = 0; i < m_iOut; ++i)  
 99 |     {  
100 |         cout << m_pdOutdata[i] << ' ';  
101 |     }  
102 |     cout << endl;  
103 |   
104 | }  
105 |   
106 |   
107 | double* NeuralBase::Forward_propagation(double* input_data)  
108 | {  
109 |     for(int i = 0; i < m_iOut; ++i)  
110 |     {  
111 |         m_pdOutdata[i] = 0.0;  
112 |         for(int j = 0; j < m_iInput; ++j)  
113 |         {  
114 |             m_pdOutdata[i] += m_ppdW[i][j]*input_data[j];  
115 |         }  
116 |         m_pdOutdata[i] += m_pdBias[i];  
117 |     }  
118 |     return m_pdOutdata;  
119 | }  
120 |   
121 | void NeuralBase::Back_propagation(double* input_data, double* pdlabel, double dLr)  
122 | {  
123 |     for(int i = 0; i < m_iOut; ++i)  
124 |     {  
125 |         m_pdDelta[i] = pdlabel[i] - m_pdOutdata[i] ;  
126 |         for(int j = 0; j < m_iInput; ++j)  
127 |         {  
128 |             m_ppdW[i][j] += dLr * m_pdDelta[i] * input_data[j] / m_iSamplenum;  
129 |         }  
130 |         m_pdBias[i] += dLr * m_pdDelta[i] / m_iSamplenum;  
131 |     }  
132 | }  
133 | void NeuralBase::MakeOneLabel(int imax, double *pdlabel)  
134 | {  
135 |     for (int j = 0; j < m_iOut; ++j)  
136 |         pdlabel[j] = 0;  
137 |     pdlabel[imax] = 1.0;  
138 | }  
139 | void NeuralBase::Writewb(const char *szName)  
140 | {  
141 |     savewb(szName, m_ppdW, m_pdBias, m_iOut, m_iInput);  
142 | }  
143 | long NeuralBase::Readwb(const char *szName, long dstartpos)  
144 | {  
145 |     return loadwb(szName, m_ppdW, m_pdBias, m_iOut, m_iInput, dstartpos);  
146 | }  
147 |   
148 | void NeuralBase::Setwb(vector<double*> &vpdw, vector<double> &vdb)  
149 | {  
150 |     assert(vpdw.size() == (DWORD)m_iOut);
151 |     for (int i = 0; i < m_iOut; ++i)  
152 |     {  
153 |         delete []m_ppdW[i];  
154 |         m_ppdW[i] = vpdw[i];  
155 |         m_pdBias[i] = vdb[i];  
156 |     }  
157 | }  
158 |   
159 | void NeuralBase::TrainAllSample(const vector<double *> &vecTrain, const vector<WORD> &vectrainlabel, double dLr)  
160 | {  
161 |     for (int j = 0; j < m_iSamplenum; ++j)  
162 |     {  
163 |         Train(vecTrain[j], vectrainlabel[j], dLr);  
164 |     }  
165 | }  
166 | void NeuralBase::Train(double *x, WORD y, double dLr)  
167 | {  
168 |     (void)x;  
169 |     (void)y;  
170 |     (void)dLr;  
171 |     cout << "NeuralBase::Train(double *x, WORD y, double dLr)" << endl;  
172 | }  
173 |   
174 | void NeuralBase::PrintOutputData()  
175 | {  
176 |     for (int i = 0; i < m_iOut; ++i)  
177 |     {  
178 |         cout << m_pdOutdata[i] << ' ';  
179 |     }  
180 |     cout << endl;  
181 |       
182 |   
183 | }  
184 | 


--------------------------------------------------------------------------------
/neuralbase.h:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/neuralbase.h


--------------------------------------------------------------------------------
/poollayer.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/poollayer.cpp


--------------------------------------------------------------------------------
/poollayer.h:
--------------------------------------------------------------------------------
 1 | #ifndef POOLLAYER_H  
 2 | #define POOLLAYER_H  
 3 | #include <vector>  
 4 | using std::vector;   
 5 | class PoolLayer  
 6 | {  
 7 | public:  
 8 |     PoolLayer(int iOutImageNumber, int iPoolWidth, int iFeatureMapWidth);  
 9 |     ~PoolLayer();  
10 |     void Convolute(double *pdInputData);  
11 |     void SetBias(const vector<double> &vecBias);  
12 |     double* GetOutputData();  
13 |     void PrintOutputData();  
14 |       
15 | private:  
16 |     int m_iOutImageNumber;  
17 |     int m_iPoolWidth;  
18 |     int m_iFeatureMapWidth;  
19 |     int m_iPoolSize;  
20 |     int m_iOutImageEdge;  
21 |     int m_iOutImageSize;  
22 |       
23 |     double *m_pdOutData;  
24 |     double *m_pdBias;  
25 | };  
26 |   
27 | #endif // POOLLAYER_H  


--------------------------------------------------------------------------------
/stdafx.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/stdafx.cpp


--------------------------------------------------------------------------------
/stdafx.h:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/stdafx.h


--------------------------------------------------------------------------------
/targetver.h:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/targetver.h


--------------------------------------------------------------------------------
/util.cpp:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/MachineLP/cnn_faceDetection/10872c30492c6dd23b713003197f378c7df68b65/util.cpp


--------------------------------------------------------------------------------
/util.h:
--------------------------------------------------------------------------------
 1 | #ifndef UTIL_H  
 2 | #define UTIL_H  
 3 | #include <iostream>  
 4 | #include <cstdio>  
 5 | #include <cstdlib>  
 6 | #include <ctime>  
 7 | #include <vector>   
 8 |   
 9 | using namespace std;  
10 |   
11 |   
12 | typedef unsigned char BYTE;  
13 | typedef unsigned short WORD;
14 | typedef unsigned int DWORD;
15 |   
16 | double sigmoid(double x);  
17 | double mytanh(double dx);  
18 |   
19 | typedef struct stShapeWb  
20 | {  
21 |     stShapeWb(int w, int h):width(w), height(h){}  
22 |     int width;  
23 |     int height;  
24 | }ShapeWb_S;  
25 |   
26 | void MakeOneLabel(int iMax, double *pdLabel, int m_iOut);  
27 |   
28 | double uniform(double _min, double _max);  
29 | //void printArr(T *parr, int num);  
30 | //void printArrDouble(double **pparr, int row, int col);  
31 | void initArr(double *parr, int num);  
32 | int getMaxIndex(double *pdarr, int num);  
33 | void Printivec(const vector<int> &ivec);  
34 | void savewb(const char *szName, double **m_ppdW, double *m_pdBias,  
35 |              int irow, int icol);  
36 | long loadwb(const char *szName, double **m_ppdW, double *m_pdBias,  
37 |             int irow, int icol, long dstartpos);  
38 |   
39 | void TestLoadJson(const char *pcfilename);  
40 | bool LoadvtFromJson(vector<double*> &vecTrain, vector<WORD> &vecLabel, const char *filename, const int m_iInput);  
41 | bool LoadwbFromJson(vector<double*> &vecTrain, vector<double> &vecLabel, const char *filename, const int m_iInput);  
42 | bool LoadTestSampleFromJson(vector<double*> &vecTrain, vector<WORD> &vecLabel, const char *filename, const int m_iInput);  
43 | bool LoadwbByByte(vector<double*> &vecTrain, vector<double> &vecLabel, const char *filename, const int m_iInput);  
44 | bool LoadallwbByByte(vector< vector<double*> > &vvAllw, vector< vector<double> > &vvAllb, const char *filename,  
45 |                      const int m_iInput, const int ihidden, const int m_iOut);  
46 | bool LoadWeighFromJson(vector< vector<double*> > &vvAllw, vector< vector<double> > &vvAllb,  
47 |                      const char *filename, const vector<int> &vecSecondDimOfWeigh);  
48 | void MakeCnnSample(double arr[2][64], double *pdImage, int iImageWidth, int iNumOfImage );  
49 | void MakeCnnWeigh(double *, int iNumOfKernel);  
50 | template <typename T>  
51 | void printArr(T *parr, int num)  
52 | {  
53 |     cout << "****printArr****" << endl;  
54 |   
55 |     for (int i = 0; i < num; ++i)  
56 |         cout << parr[i] << ' ';  
57 |     cout << endl;  
58 | }  
59 | template <typename T>  
60 | void printArrDouble(T **pparr, int row, int col)  
61 | {  
62 |     cout << "****printArrDouble****" << endl;  
63 |     for (int i = 0; i < row; ++i)  
64 |     {  
65 |         for (int j = 0; j < col; ++j)  
66 |         {  
67 |             cout << pparr[i][j] << ' ';  
68 |         }  
69 |         cout << endl;  
70 |     }  
71 | }  
72 |   
73 |   
74 | #endif  
75 | 


--------------------------------------------------------------------------------