├── Code ├── GUIApp.py ├── SortTracker.py ├── classify_track_count.py ├── extras │ ├── coco.data │ ├── coco.names │ ├── multi_classify.cfg │ ├── multi_classify.data │ ├── multi_classify.names │ └── yolov4.cfg ├── media │ └── Pics_Readme │ │ ├── GUI.png │ │ └── fulldemo.gif └── requirements.txt ├── LICENSE ├── README.md └── setup.bash /Code/GUIApp.py: -------------------------------------------------------------------------------- 1 | from PyQt5 import QtCore, QtGui, QtWidgets 2 | import cv2 3 | from classify_track_count import * 4 | import os 5 | 6 | 7 | class Ui_MainWindow(object): 8 | def setupUi(self, MainWindow): 9 | MainWindow.setObjectName("MainWindow") 10 | MainWindow.resize(800, 600) 11 | MainWindow.setWindowIcon(QtGui.QIcon("logo.png")) 12 | self.centralwidget = QtWidgets.QWidget(MainWindow) 13 | self.centralwidget.setObjectName("centralwidget") 14 | self.gridLayoutWidget = QtWidgets.QWidget(self.centralwidget) 15 | self.gridLayoutWidget.setGeometry(QtCore.QRect(100, 120, 600, 250)) 16 | self.gridLayoutWidget.setObjectName("gridLayoutWidget") 17 | self.gridLayout = QtWidgets.QGridLayout(self.gridLayoutWidget) 18 | self.gridLayout.setContentsMargins(0, 0, 0, 20) 19 | self.gridLayout.setObjectName("gridLayout") 20 | self.progress = QtWidgets.QProgressBar(self.centralwidget) 21 | self.progress.setGeometry(QtCore.QRect(300, 330, 211, 21)) 22 | self.progress.setProperty("value", 0) 23 | self.progress.setObjectName("progress") 24 | self.label2 = QtWidgets.QLabel(self.gridLayoutWidget) 25 | font = QtGui.QFont() 26 | font.setPointSize(10) 27 | self.label2.setFont(font) 28 | self.label2.setObjectName("label2") 29 | self.gridLayout.addWidget(self.label2, 2, 0, 1, 1, QtCore.Qt.AlignHCenter) 30 | self.button_roi = QtWidgets.QPushButton(self.gridLayoutWidget) 31 | font = QtGui.QFont() 32 | font.setPointSize(10) 33 | self.button_roi.setFont(font) 34 | self.button_roi.setCursor(QtGui.QCursor(QtCore.Qt.PointingHandCursor)) 35 | self.button_roi.setObjectName("button_roi") 36 | self.gridLayout.addWidget(self.button_roi, 2, 2, 1, 1) 37 | self.button_upload = QtWidgets.QPushButton(self.gridLayoutWidget) 38 | font = QtGui.QFont() 39 | font.setPointSize(10) 40 | self.button_upload.setFont(font) 41 | self.button_upload.setCursor(QtGui.QCursor(QtCore.Qt.PointingHandCursor)) 42 | self.button_upload.setObjectName("button_upload") 43 | self.gridLayout.addWidget(self.button_upload, 0, 2, 1, 1) 44 | self.input2 = QtWidgets.QLineEdit(self.gridLayoutWidget) 45 | font = QtGui.QFont() 46 | font.setPointSize(10) 47 | self.input2.setFont(font) 48 | self.input2.setObjectName("input2") 49 | self.gridLayout.addWidget(self.input2, 2, 1, 1, 1) 50 | self.input1 = QtWidgets.QLineEdit(self.gridLayoutWidget) 51 | font = QtGui.QFont() 52 | font.setPointSize(10) 53 | self.input1.setFont(font) 54 | self.input1.setObjectName("input1") 55 | self.gridLayout.addWidget(self.input1, 0, 1, 1, 1) 56 | self.label1 = QtWidgets.QLabel(self.gridLayoutWidget) 57 | font = QtGui.QFont() 58 | font.setPointSize(10) 59 | self.label1.setFont(font) 60 | self.label1.setObjectName("label1") 61 | self.gridLayout.addWidget(self.label1, 0, 0, 1, 1) 62 | self.start = QtWidgets.QPushButton(self.gridLayoutWidget) 63 | font = QtGui.QFont() 64 | font.setPointSize(10) 65 | self.start.setFont(font) 66 | self.start.setCursor(QtGui.QCursor(QtCore.Qt.PointingHandCursor)) 67 | self.start.setObjectName("start") 68 | self.gridLayout.addWidget(self.start, 3, 1, 1, 1) 69 | self.show_results = QtWidgets.QPushButton(self.centralwidget) 70 | self.show_results.setGeometry(QtCore.QRect(240, 450, 320, 40)) 71 | font = QtGui.QFont() 72 | font.setPointSize(10) 73 | self.show_results.setFont(font) 74 | self.show_results.setCursor(QtGui.QCursor(QtCore.Qt.PointingHandCursor)) 75 | self.show_results.setObjectName("show_results") 76 | self.show_results.hide() 77 | self.loading = QtWidgets.QLabel(self.centralwidget) 78 | self.loading.setGeometry(QtCore.QRect(290, 400, 320, 40)) 79 | font.setPointSize(14) 80 | self.loading.setFont(font) 81 | self.loading.setObjectName("loading") 82 | self.loading.hide() 83 | MainWindow.setCentralWidget(self.centralwidget) 84 | self.menubar = QtWidgets.QMenuBar(MainWindow) 85 | self.menubar.setGeometry(QtCore.QRect(0, 0, 800, 18)) 86 | self.menubar.setObjectName("menubar") 87 | MainWindow.setMenuBar(self.menubar) 88 | self.statusbar = QtWidgets.QStatusBar(MainWindow) 89 | self.statusbar.setObjectName("statusbar") 90 | MainWindow.setStatusBar(self.statusbar) 91 | 92 | self.retranslateUi(MainWindow) 93 | QtCore.QMetaObject.connectSlotsByName(MainWindow) 94 | 95 | self.button_upload.clicked.connect(self.upload_video) 96 | self.button_roi.clicked.connect(self.select_roi) 97 | self.start.clicked.connect(self.run_darknet) 98 | self.show_results.clicked.connect(self.open_file) 99 | 100 | def retranslateUi(self, MainWindow): 101 | _translate = QtCore.QCoreApplication.translate 102 | MainWindow.setWindowTitle(_translate("MainWindow", "Vehicle Classifier")) 103 | self.label2.setText(_translate("MainWindow", "ROI selected: ")) 104 | self.button_roi.setText(_translate("MainWindow", "Choose ROI")) 105 | self.button_upload.setText(_translate("MainWindow", "Upload video")) 106 | self.input2.setText(_translate("MainWindow", "")) 107 | self.input1.setText(_translate("MainWindow", "")) 108 | self.label1.setText(_translate("MainWindow", "File Location: ")) 109 | self.start.setText(_translate("MainWindow", "START")) 110 | self.loading.setText(_translate("MainWindow", "This might take a while...")) 111 | self.show_results.setText(_translate("MainWindow", "Show Results")) 112 | 113 | def clear_field(self): 114 | self.input1.setText("") 115 | self.input2.setText("") 116 | 117 | def upload_video(self): 118 | options = QtWidgets.QFileDialog.Options() 119 | options |= QtWidgets.QFileDialog.DontUseNativeDialog 120 | fileName, _ = QtWidgets.QFileDialog.getOpenFileName( 121 | None, 122 | "Upload Video File", 123 | "", 124 | "All Files (*);;Video Files (*.mp4);;Video Files (*.avi)", 125 | options=options) 126 | if fileName: 127 | self.input1.setText(fileName) 128 | self.fileName = fileName 129 | 130 | def select_roi(self): 131 | try: 132 | # Mouse event function 133 | def click_event(event, x, y, flags, param): 134 | if event == cv2.EVENT_LBUTTONDOWN: 135 | self.points.append((x,y)) 136 | if event == cv2.EVENT_LBUTTONUP: 137 | self.points.append((x,y)) 138 | print(self.points) 139 | #if len(self.points)%2 == 2: 140 | self.p1 = self.points[len(self.points)-2] 141 | self.p2 = self.points[len(self.points)-1] 142 | cv2.rectangle(img, self.p1, self.p2,(0,255,0), 2 ) 143 | self.input2.setText("("+str(self.p1[0])+","+str(self.p1[1])+")"+ ", "+"("+str(self.p2[0])+","+str(self.p2[1])+")") 144 | #points.clear() 145 | #print(p1 , p2) 146 | #cv2.imshow("frame", img) 147 | 148 | 149 | cap = cv2.VideoCapture(self.fileName) 150 | _, img = cap.read() 151 | cv2.putText(img, "Click and drag to select ROI", (20,30), cv2.FONT_HERSHEY_SIMPLEX, 1, (255,0,0), 2) 152 | cv2.putText(img, "Click 'Enter' to Proceed", (20,60), cv2.FONT_HERSHEY_SIMPLEX, 1, (255,0,0), 2) 153 | cv2.imshow("frame", img) 154 | self.points = [] 155 | cv2.setMouseCallback("frame", click_event) 156 | key = cv2.waitKey(0) 157 | if key == ord('q'): 158 | cv2.destroyAllWindows() 159 | 160 | except: 161 | msg = QtWidgets.QMessageBox() 162 | msg.setWindowTitle("Error") 163 | msg.setWindowIcon(QtGui.QIcon("logo.png")) 164 | msg.setText("File type is not supported") 165 | msg.setIcon(QtWidgets.QMessageBox.Warning) 166 | msg.buttonClicked.connect(self.clear_field) 167 | x = msg.exec_() 168 | 169 | def run_darknet(self): 170 | self.loading.show() 171 | self.results, self.file_loc = YOLO(self.p1 , self.p2 , self.progress , self.fileName)#str(self.input1.text()), str(self.input2.text())) 172 | if self.results: 173 | self.loading.hide() 174 | self.show_results.show() 175 | print("Results obtained!!") 176 | 177 | def open_file(self): 178 | os.startfile(self.file_loc) 179 | 180 | 181 | if __name__ == "__main__": 182 | import sys 183 | app = QtWidgets.QApplication(sys.argv) 184 | MainWindow = QtWidgets.QMainWindow() 185 | ui = Ui_MainWindow() 186 | ui.setupUi(MainWindow) 187 | MainWindow.show() 188 | sys.exit(app.exec_()) 189 | -------------------------------------------------------------------------------- /Code/SortTracker.py: -------------------------------------------------------------------------------- 1 | """ 2 | SORT: A Simple, Online and Realtime Tracker 3 | Copyright (C) 2016-2020 Alex Bewley alex@bewley.ai 4 | This program is free software: you can redistribute it and/or modify 5 | it under the terms of the GNU General Public License as published by 6 | the Free Software Foundation, either version 3 of the License, or 7 | (at your option) any later version. 8 | This program is distributed in the hope that it will be useful, 9 | but WITHOUT ANY WARRANTY; without even the implied warranty of 10 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 11 | GNU General Public License for more details. 12 | You should have received a copy of the GNU General Public License 13 | along with this program. If not, see . 14 | """ 15 | from __future__ import print_function 16 | 17 | import os 18 | import numpy as np 19 | import matplotlib 20 | matplotlib.use('TkAgg') 21 | import matplotlib.pyplot as plt 22 | import matplotlib.patches as patches 23 | from skimage import io 24 | 25 | import glob 26 | import time 27 | import argparse 28 | from filterpy.kalman import KalmanFilter 29 | 30 | try: 31 | from numba import jit 32 | except: 33 | def jit(func): 34 | return func 35 | 36 | np.random.seed(0) 37 | 38 | 39 | def linear_assignment(cost_matrix): 40 | try: 41 | import lap 42 | _, x, y = lap.lapjv(cost_matrix, extend_cost=True) 43 | return np.array([[y[i],i] for i in x if i >= 0]) # 44 | except ImportError: 45 | from scipy.optimize import linear_sum_assignment 46 | x, y = linear_sum_assignment(cost_matrix) 47 | return np.array(list(zip(x, y))) 48 | 49 | 50 | @jit 51 | def iou(bb_test, bb_gt): 52 | # current frame(detections) , past frame (trackers) 53 | # x increases as we go to the right and y increases as we go down 54 | # iou = intersection/(rect1 + rect2 - intersection) 55 | """ 56 | Computes IUO between two bboxes in the form [x1,y1,x2,y2] 57 | """ 58 | xx1 = np.maximum(bb_test[0], bb_gt[0]) 59 | yy1 = np.maximum(bb_test[1], bb_gt[1]) 60 | xx2 = np.minimum(bb_test[2], bb_gt[2]) 61 | yy2 = np.minimum(bb_test[3], bb_gt[3]) 62 | w = np.maximum(0., xx2 - xx1) 63 | h = np.maximum(0., yy2 - yy1) 64 | wh = w * h 65 | o = wh / ((bb_test[2] - bb_test[0]) * (bb_test[3] - bb_test[1]) 66 | + (bb_gt[2] - bb_gt[0]) * (bb_gt[3] - bb_gt[1]) - wh) 67 | return(o) 68 | 69 | 70 | def convert_bbox_to_z(bbox): 71 | """ 72 | Takes a bounding box in the form [x1,y1,x2,y2] and returns z in the form 73 | [x,y,s,r] where x,y is the centre of the box and s is the scale/area and r is 74 | the aspect ratio 75 | """ 76 | w = bbox[2] - bbox[0] 77 | h = bbox[3] - bbox[1] 78 | x = bbox[0] + w/2. 79 | y = bbox[1] + h/2. 80 | s = w * h #scale is just area 81 | r = w / float(h) 82 | return np.array([x, y, s, r]).reshape((4, 1)) 83 | 84 | 85 | def convert_x_to_bbox(x,score=None): 86 | """ 87 | Takes a bounding box in the centre form [x,y,s,r] and returns it in the form 88 | [x1,y1,x2,y2] where x1,y1 is the top left and x2,y2 is the bottom right 89 | """ 90 | w = np.sqrt(x[2] * x[3]) 91 | h = x[2] / w 92 | if(score==None): 93 | return np.array([x[0]-w/2.,x[1]-h/2.,x[0]+w/2.,x[1]+h/2.]).reshape((1,4)) 94 | else: 95 | return np.array([x[0]-w/2.,x[1]-h/2.,x[0]+w/2.,x[1]+h/2.,score]).reshape((1,5)) 96 | 97 | 98 | class KalmanBoxTracker(object): 99 | """ 100 | This class represents the internal state of individual tracked objects observed as bbox. 101 | """ 102 | count = 0 103 | def __init__(self,bbox): 104 | """ 105 | Initialises a tracker using initial bounding box. 106 | """ 107 | #define constant velocity model 108 | self.kf = KalmanFilter(dim_x=7, dim_z=4) 109 | self.kf.F = np.array([[1,0,0,0,1,0,0],[0,1,0,0,0,1,0],[0,0,1,0,0,0,1],[0,0,0,1,0,0,0], [0,0,0,0,1,0,0],[0,0,0,0,0,1,0],[0,0,0,0,0,0,1]]) 110 | self.kf.H = np.array([[1,0,0,0,0,0,0],[0,1,0,0,0,0,0],[0,0,1,0,0,0,0],[0,0,0,1,0,0,0]]) 111 | 112 | self.kf.R[2:,2:] *= 10. 113 | self.kf.P[4:,4:] *= 1000. #give high uncertainty to the unobservable initial velocities 114 | self.kf.P *= 10. 115 | self.kf.Q[-1,-1] *= 0.01 116 | self.kf.Q[4:,4:] *= 0.01 117 | 118 | self.kf.x[:4] = convert_bbox_to_z(bbox) 119 | self.time_since_update = 0 120 | self.id = KalmanBoxTracker.count 121 | KalmanBoxTracker.count += 1 122 | self.history = [] 123 | self.hits = 0 124 | self.hit_streak = 0 125 | self.age = 0 126 | 127 | def update(self,bbox): 128 | """ 129 | Updates the state vector with observed bbox. 130 | """ 131 | self.time_since_update = 0 132 | self.history = [] 133 | self.hits += 1 134 | self.hit_streak += 1 135 | self.kf.update(convert_bbox_to_z(bbox)) 136 | 137 | def predict(self): 138 | """ 139 | Advances the state vector and returns the predicted bounding box estimate. 140 | """ 141 | if((self.kf.x[6]+self.kf.x[2])<=0): 142 | self.kf.x[6] *= 0.0 143 | self.kf.predict() 144 | self.age += 1 145 | if(self.time_since_update>0): 146 | self.hit_streak = 0 147 | self.time_since_update += 1 148 | self.history.append(convert_x_to_bbox(self.kf.x)) 149 | return self.history[-1] 150 | 151 | def get_state(self): 152 | """ 153 | Returns the current bounding box estimate. 154 | """ 155 | return convert_x_to_bbox(self.kf.x) 156 | 157 | 158 | def associate_detections_to_trackers(detections,trackers,iou_threshold = 0.3): 159 | """ 160 | Assigns detections to tracked object (both represented as bounding boxes) 161 | Returns 3 lists of matches, unmatched_detections and unmatched_trackers 162 | """ 163 | if(len(trackers)==0): 164 | return np.empty((0,2),dtype=int), np.arange(len(detections)), np.empty((0,5),dtype=int) 165 | iou_matrix = np.zeros((len(detections),len(trackers)),dtype=np.float32) 166 | 167 | for d,det in enumerate(detections): 168 | for t,trk in enumerate(trackers): 169 | iou_matrix[d,t] = iou(det,trk) 170 | 171 | if min(iou_matrix.shape) > 0: 172 | a = (iou_matrix > iou_threshold).astype(np.int32) 173 | if a.sum(1).max() == 1 and a.sum(0).max() == 1: 174 | matched_indices = np.stack(np.where(a), axis=1) 175 | else: 176 | matched_indices = linear_assignment(-iou_matrix) 177 | else: 178 | matched_indices = np.empty(shape=(0,2)) 179 | 180 | unmatched_detections = [] 181 | for d, det in enumerate(detections): 182 | # for unmatched dets we use matched_indices[:,0] because matched indices store associations found as [detection , tracker] 183 | if(d not in matched_indices[:,0]): 184 | unmatched_detections.append(d) 185 | unmatched_trackers = [] 186 | for t, trk in enumerate(trackers): 187 | if(t not in matched_indices[:,1]): 188 | unmatched_trackers.append(t) 189 | 190 | #filter out matched with low IOU 191 | matches = [] 192 | for m in matched_indices: 193 | if(iou_matrix[m[0], m[1]]= self.min_hits or self.frame_count <= self.min_hits): 256 | ret.append(np.concatenate((d,[trk.id+1])).reshape(1,-1)) # +1 as MOT benchmark requires positive 257 | i -= 1 258 | # remove dead tracklet 259 | if(trk.time_since_update > self.max_age): 260 | self.trackers.pop(i) 261 | if(len(ret)>0): 262 | return np.concatenate(ret) 263 | return np.empty((0,5)) 264 | 265 | def parse_args(): 266 | """Parse input arguments.""" 267 | parser = argparse.ArgumentParser(description='SORT demo') 268 | parser.add_argument('--display', dest='display', help='Display online tracker output (slow) [False]',action='store_true') 269 | parser.add_argument("--seq_path", help="Path to detections.", type=str, default='data') 270 | parser.add_argument("--phase", help="Subdirectory in seq_path.", type=str, default='train') 271 | args = parser.parse_args() 272 | return args 273 | 274 | if __name__ == '__main__': 275 | # all train 276 | args = parse_args() 277 | display = args.display 278 | phase = args.phase 279 | total_time = 0.0 280 | total_frames = 0 281 | colours = np.random.rand(32, 3) #used only for display 282 | if(display): 283 | if not os.path.exists('mot_benchmark'): 284 | print('\n\tERROR: mot_benchmark link not found!\n\n Create a symbolic link to the MOT benchmark\n (https://motchallenge.net/data/2D_MOT_2015/#download). E.g.:\n\n $ ln -s /path/to/MOT2015_challenge/2DMOT2015 mot_benchmark\n\n') 285 | exit() 286 | plt.ion() 287 | fig = plt.figure() 288 | ax1 = fig.add_subplot(111, aspect='equal') 289 | 290 | if not os.path.exists('output'): 291 | os.makedirs('output') 292 | pattern = os.path.join(args.seq_path, phase, '*', 'det', 'det.txt') 293 | for seq_dets_fn in glob.glob(pattern): 294 | mot_tracker = Sort() #create instance of the SORT tracker 295 | seq_dets = np.loadtxt(seq_dets_fn, delimiter=',') 296 | seq = seq_dets_fn[pattern.find('*'):].split('/')[0] 297 | 298 | with open('output/%s.txt'%(seq),'w') as out_file: 299 | print("Processing %s."%(seq)) 300 | for frame in range(int(seq_dets[:,0].max())): 301 | frame += 1 #detection and frame numbers begin at 1 302 | dets = seq_dets[seq_dets[:, 0]==frame, 2:7] 303 | dets[:, 2:4] += dets[:, 0:2] #convert to [x1,y1,w,h] to [x1,y1,x2,y2] 304 | total_frames += 1 305 | 306 | if(display): 307 | fn = 'mot_benchmark/%s/%s/img1/%06d.jpg'%(phase, seq, frame) 308 | im =io.imread(fn) 309 | ax1.imshow(im) 310 | plt.title(seq + ' Tracked Targets') 311 | 312 | start_time = time.time() 313 | trackers = mot_tracker.update(dets) 314 | cycle_time = time.time() - start_time 315 | total_time += cycle_time 316 | 317 | for d in trackers: 318 | print('%d,%d,%.2f,%.2f,%.2f,%.2f,1,-1,-1,-1'%(frame,d[4],d[0],d[1],d[2]-d[0],d[3]-d[1]),file=out_file) 319 | if(display): 320 | d = d.astype(np.int32) 321 | ax1.add_patch(patches.Rectangle((d[0],d[1]),d[2]-d[0],d[3]-d[1],fill=False,lw=3,ec=colours[d[4]%32,:])) 322 | 323 | if(display): 324 | fig.canvas.flush_events() 325 | plt.draw() 326 | ax1.cla() 327 | 328 | print("Total Tracking took: %.3f seconds for %d frames or %.1f FPS" % (total_time, total_frames, total_frames / total_time)) 329 | 330 | if(display): 331 | print("Note: to get real runtime results run without the option: --display") 332 | -------------------------------------------------------------------------------- /Code/classify_track_count.py: -------------------------------------------------------------------------------- 1 | from ctypes import * 2 | import math 3 | import random 4 | import os 5 | import cv2 6 | import numpy as np 7 | import time 8 | import darknet 9 | import csv 10 | import time 11 | 12 | # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>. 13 | from SortTracker import * 14 | tracker = Sort() 15 | memory = {} 16 | counter = 0 17 | # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>. 18 | 19 | #>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 20 | 21 | # Return true if line segments AB and CD intersect 22 | def intersect(A,B,C,D): 23 | return ccw(A,C,D) != ccw(B,C,D) and ccw(A,B,C) != ccw(A,B,D) 24 | 25 | def ccw(A,B,C): 26 | return (C[1]-A[1]) * (B[0]-A[0]) > (B[1]-A[1]) * (C[0]-A[0]) 27 | 28 | #>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 29 | 30 | 31 | def convertBack(x, y, w, h): 32 | xmin = int(round(x - (w / 2))) 33 | xmax = int(round(x + (w / 2))) 34 | ymin = int(round(y - (h / 2))) 35 | ymax = int(round(y + (h / 2))) 36 | return xmin, ymin, xmax, ymax 37 | 38 | 39 | def cvDrawBoxes(detections, img): 40 | for detection in detections: 41 | x, y, w, h = detection[2][0],\ 42 | detection[2][1],\ 43 | detection[2][2],\ 44 | detection[2][3] 45 | xmin, ymin, xmax, ymax = convertBack( 46 | float(x), float(y), float(w), float(h)) 47 | pt1 = (xmin, ymin) 48 | pt2 = (xmax, ymax) 49 | if (detection[1] * 100) > 90: 50 | cv2.rectangle(img, pt1, pt2, (0, 255, 0), 1) 51 | cv2.putText(img, 52 | detection[0].decode() + 53 | " [" + str(round(detection[1] * 100, 2)) + "]", 54 | (pt1[0], pt1[1] - 5), cv2.FONT_HERSHEY_SIMPLEX, 0.5, 55 | [0, 255, 0], 2) 56 | return img 57 | 58 | netMain = None 59 | metaMain = None 60 | altNames = None 61 | 62 | netMain1 = None 63 | metaMain1 = None 64 | altNames1 = None 65 | 66 | car_cnt = 0 67 | truck_cnt = 0 68 | motorbike_cnt = 0 69 | bus_cnt = 0 70 | entry = {} 71 | speed = 0 72 | time_last = 0 73 | last_count = 0 74 | count_inc = 0 75 | time_interval = 0 76 | twoaxel = 0 77 | threeaxel = 0 78 | fouraxel = 0 79 | fiveaxel = 0 80 | sixaxel = 0 81 | cur_frame = 0 82 | 83 | def YOLO(roipt1 , roipt2 , progressObj , filename): 84 | global metaMain, netMain, altNames , metaMain1 , netMain1 , altNames1 , car_cnt , truck_cnt , motorbike_cnt , bus_cnt , time_last , last_count , count_inc , cur_frame , time_interval , memory , counter , twoaxel, threeaxel, fouraxel, fiveaxel, sixaxel 85 | configPath = "./cfg/yolov4.cfg" 86 | weightPath = "./yolov4.weights" 87 | metaPath = "./cfg/coco.data" 88 | 89 | configPath1 = "./cfg/multi_classify.cfg" 90 | weightPath1 = "./weights/multiclassify_4000.weights" 91 | metaPath1 = "./data/multi_classify.data" 92 | 93 | if not os.path.exists(configPath): 94 | raise ValueError("Invalid config path `" + 95 | os.path.abspath(configPath)+"`") 96 | if not os.path.exists(weightPath): 97 | raise ValueError("Invalid weight path `" + 98 | os.path.abspath(weightPath)+"`") 99 | if not os.path.exists(metaPath): 100 | raise ValueError("Invalid data file path `" + 101 | os.path.abspath(metaPath)+"`") 102 | if netMain is None: 103 | netMain = darknet.load_net_custom(configPath.encode( 104 | "ascii"), weightPath.encode("ascii"), 0, 1) # batch size = 1 105 | if metaMain is None: 106 | metaMain = darknet.load_meta(metaPath.encode("ascii")) 107 | if altNames is None: 108 | try: 109 | with open(metaPath) as metaFH: 110 | metaContents = metaFH.read() 111 | import re 112 | match = re.search("names *= *(.*)$", metaContents, 113 | re.IGNORECASE | re.MULTILINE) 114 | if match: 115 | result = match.group(1) 116 | else: 117 | result = None 118 | try: 119 | if os.path.exists(result): 120 | with open(result) as namesFH: 121 | namesList = namesFH.read().strip().split("\n") 122 | altNames = [x.strip() for x in namesList] 123 | except TypeError: 124 | pass 125 | except Exception: 126 | pass 127 | 128 | 129 | 130 | if not os.path.exists(configPath1): 131 | raise ValueError("Invalid config path `" + 132 | os.path.abspath(configPath1)+"`") 133 | if not os.path.exists(weightPath1): 134 | raise ValueError("Invalid weight path `" + 135 | os.path.abspath(weightPath1)+"`") 136 | if not os.path.exists(metaPath1): 137 | raise ValueError("Invalid data file path `" + 138 | os.path.abspath(metaPath1)+"`") 139 | if netMain1 is None: 140 | netMain1 = darknet.load_net_custom(configPath1.encode( 141 | "ascii"), weightPath1.encode("ascii"), 0, 1) # batch size = 1 142 | if metaMain1 is None: 143 | metaMain1 = darknet.load_meta(metaPath1.encode("ascii")) 144 | if altNames1 is None: 145 | try: 146 | with open(metaPath1) as metaFH: 147 | metaContents = metaFH.read() 148 | import re 149 | match = re.search("names *= *(.*)$", metaContents, 150 | re.IGNORECASE | re.MULTILINE) 151 | if match: 152 | result = match.group(1) 153 | else: 154 | result = None 155 | try: 156 | if os.path.exists(result): 157 | with open(result) as namesFH: 158 | namesList = namesFH.read().strip().split("\n") 159 | altNames1 = [x.strip() for x in namesList] 160 | except TypeError: 161 | pass 162 | except Exception: 163 | pass 164 | 165 | 166 | cap = cv2.VideoCapture(filename) 167 | print(cap.get(cv2.CAP_PROP_FPS)) 168 | tot_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT)) 169 | cap.set(3, 1280) 170 | cap.set(4, 720) 171 | out = cv2.VideoWriter( 172 | "axelCount_output.avi", cv2.VideoWriter_fourcc(*"MJPG"), 10.0, 173 | (darknet.network_width(netMain), darknet.network_height(netMain))) 174 | print("Starting the YOLO loop...") 175 | 176 | # Create an image we reuse for each detect 177 | darknet_image = darknet.make_image(darknet.network_width(netMain), 178 | darknet.network_height(netMain),3) 179 | 180 | darknet_image1 = darknet.make_image(darknet.network_width(netMain), 181 | darknet.network_height(netMain),3) 182 | 183 | while True: 184 | print(cur_frame*100/tot_frames) 185 | progressObj.setProperty("value", cur_frame*100/tot_frames) 186 | prev_time = time.time() 187 | ret, frame_read = cap.read() 188 | cur_frame+=1 189 | if ret: 190 | ROI = np.copy(frame_read) 191 | ROI[: , : , :] = 0 192 | reqdRegion = np.copy(frame_read) 193 | reqdRegion[: , : , :] = 0 194 | roipi1x = roipt1[0] 195 | roipt1y = roipt1[1] 196 | roipt2x = roipt2[0] 197 | roipt2y = roipt2[1] 198 | #region = frame_read[240:790 , 400:1400] 199 | #ROI[240:790 , 400:1400] = region 200 | #cv2.rectangle(frame_read , (400 , 240) , (1400 , 790) , (173 , 50, 200) , 2) 201 | region = frame_read[roipt1y:roipt2y , roipi1x:roipt2x] 202 | ROI[roipt1y:roipt2y , roipi1x:roipt2x] = region 203 | cv2.rectangle(frame_read , roipt1 , roipt2 , (173 , 50, 200) , 2) 204 | frame_rgb = cv2.cvtColor(frame_read, cv2.COLOR_BGR2RGB) 205 | 206 | frame_resized = cv2.resize(frame_rgb, 207 | (darknet.network_width(netMain), 208 | darknet.network_height(netMain)), 209 | interpolation=cv2.INTER_LINEAR) 210 | ROI_resized = cv2.resize(ROI, 211 | (darknet.network_width(netMain), 212 | darknet.network_height(netMain)), 213 | interpolation=cv2.INTER_LINEAR) 214 | reqdRegion_resized = cv2.resize(reqdRegion, 215 | (darknet.network_width(netMain), 216 | darknet.network_height(netMain)), 217 | interpolation=cv2.INTER_LINEAR) 218 | 219 | darknet.copy_image_from_bytes(darknet_image , ROI_resized.tobytes()) 220 | 221 | detections = darknet.detect_image(netMain, metaMain, darknet_image, thresh=0.25) 222 | image = cvDrawBoxes(detections, frame_resized) 223 | image_clean = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) 224 | image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) 225 | 226 | # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>. 227 | dets = [] 228 | full_dets=[] 229 | if len(detections) > 0: 230 | # loop over the indexes we are keeping 231 | for i in range (0,len(detections)): 232 | if detections[i][0].decode() != 'person': 233 | #print(len(detections)) 234 | (x, y) = (detections[i][2][0], detections[i][2][1]) 235 | (w, h) = (detections[i][2][2] , detections[i][2][3] ) 236 | dets.append([float(x-w/2), float(y-h/2), float(x+w/2), float(y+h/2), float(detections[i][1])]) 237 | full_dets.append([int((float(x-w/2)+float(x+w/2))/2) , int((float(y-h/2)+float(y+h/2))/2), detections[i][0].decode()]) 238 | #print(np.shape(dets)) 239 | #print(full_dets) 240 | 241 | print(full_dets) 242 | print('>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>') 243 | dets = np.asarray(dets) 244 | print(dets) 245 | print("##################################") 246 | tracks = tracker.update(dets) 247 | #print(tracks) 248 | 249 | boxes = [] 250 | indexIDs = [] 251 | c = [] 252 | previous = memory.copy() 253 | memory = {} 254 | 255 | for track in tracks: 256 | # As boxes co-ordinates and indexes are appended one by one we are storing them 257 | # in a dictionary 258 | #print(track) 259 | for i , det_cent in enumerate(full_dets): 260 | euclidean_distance = math.sqrt( (det_cent[0]-(track[0]+track[2])/2)**2 + (det_cent[1]-(track[1]+track[3])/2)**2 ) 261 | if euclidean_distance <= 4: 262 | boxes.append([track[0], track[1], track[2], track[3], det_cent[2]]) 263 | indexIDs.append(int(track[4])) 264 | memory[indexIDs[-1]] = boxes[-1] 265 | break 266 | 267 | if len(boxes) > 0: 268 | i = int(0) 269 | for box in boxes: 270 | # extract the bounding box coordinates 271 | (x, y) = (int(box[0]), int(box[1])) 272 | (w, h) = (int(box[2]), int(box[3])) 273 | cv2.rectangle(image, (int(x), int(y)), (int(w), int(h)), (255 , 0 , 0), 2) 274 | 275 | if indexIDs[i] in previous: 276 | previous_box = previous[indexIDs[i]] 277 | (x2, y2) = (int(previous_box[0]), int(previous_box[1])) 278 | (w2, h2) = (int(previous_box[2]), int(previous_box[3])) 279 | # p0 = prevCentroid 280 | # p1 = current centroid 281 | # condition for counter to increment is if line between prev centroid and current centroid intersect then increment counter 282 | p0 = (int(x + (w-x)/2), int(y + (h-y)/2)) 283 | p1 = (int(x2 + (w2-x2)/2), int(y2 + (h2-y2)/2)) 284 | cv2.putText(image, str(indexIDs[i]), p0 , cv2.FONT_HERSHEY_SIMPLEX , 0.5,(0, 255, 0) , 2) 285 | cv2.line(image , p0, p1, (0 , 255 , 0), 2) 286 | 287 | countwheel=0 288 | # If diagonal intersects with line then increment counter 289 | if intersect(p0, p1, (340 , 366) , (340 , 138)): 290 | print('entered') 291 | 292 | if box[4] == 'car': 293 | car_cnt+=1 294 | if box[4] == 'truck': 295 | print("Counting axels...........") 296 | region = image_clean[y-20:h+20 , x-20:w+20] 297 | reqdRegion_resized[y-20:h+20 , x-20:w+20] = region 298 | reqdRegion_resized = cv2.cvtColor(reqdRegion_resized , cv2.COLOR_RGB2BGR) 299 | darknet.copy_image_from_bytes(darknet_image1 , reqdRegion_resized.tobytes()) 300 | detections = darknet.detect_image(netMain1, metaMain1, darknet_image1, thresh=0.25) 301 | reqdRegion_resized = cv2.cvtColor(reqdRegion_resized , cv2.COLOR_BGR2RGB) 302 | reqdRegion_resized = cvDrawBoxes(detections, reqdRegion_resized) 303 | if len(detections) > 0: 304 | for j in range (0,len(detections)): 305 | if detections[j][0].decode() == 'wheel': 306 | countwheel+=1 307 | print('No of wheeles are: ' , countwheel) 308 | #cv2.imshow("img" , reqdRegion_resized) 309 | #cv2.waitKey(0) 310 | print("Done counting..........") 311 | truck_cnt+=1 312 | if countwheel == 2: 313 | twoaxel+=1 314 | if countwheel == 3: 315 | threeaxel+=1 316 | if countwheel == 4: 317 | fouraxel+=1 318 | if countwheel == 5: 319 | fiveaxel+=1 320 | if countwheel == 6: 321 | sixaxel+=1 322 | if box[4] == 'motorbike': 323 | motorbike_cnt+=1 324 | if box[4] == 'bus': 325 | bus_cnt+=1 326 | counter += 1 327 | i+=1 328 | cv2.putText(image, 329 | "Total Count = {}".format(counter), 330 | (10 , 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, 331 | [0, 255, 0], 2) 332 | cv2.putText(image, 333 | "Car Count = {}".format(car_cnt), 334 | (10 , 30), cv2.FONT_HERSHEY_SIMPLEX, 0.5, 335 | [0, 255, 0], 2) 336 | cv2.putText(image, 337 | "Truck Count = {}".format(truck_cnt), 338 | (10 , 50), cv2.FONT_HERSHEY_SIMPLEX, 0.5, 339 | [0, 255, 0], 2) 340 | cv2.putText(image, 341 | "2-Axel Truck Count = {}".format(twoaxel), 342 | (20 , 70), cv2.FONT_HERSHEY_SIMPLEX, 0.4, 343 | [0, 255, 0], 2) 344 | cv2.putText(image, 345 | "3-Axel Truck Count = {}".format(threeaxel), 346 | (20 , 90), cv2.FONT_HERSHEY_SIMPLEX, 0.4, 347 | [0, 255, 0], 2) 348 | cv2.putText(image, 349 | "4-Axel Truck Count = {}".format(fouraxel), 350 | (20 , 110), cv2.FONT_HERSHEY_SIMPLEX, 0.4, 351 | [0, 255, 0], 2) 352 | cv2.putText(image, 353 | "5-Axel Truck Count = {}".format(fiveaxel), 354 | (20 , 130), cv2.FONT_HERSHEY_SIMPLEX, 0.4, 355 | [0, 255, 0], 2) 356 | cv2.putText(image, 357 | "6-Axel Truck Count = {}".format(sixaxel), 358 | (20 , 150), cv2.FONT_HERSHEY_SIMPLEX, 0.4, 359 | [0, 255, 0], 2) 360 | cv2.putText(image, 361 | "Motorbike Count = {}".format(motorbike_cnt), 362 | (10 , 170), cv2.FONT_HERSHEY_SIMPLEX, 0.5, 363 | [0, 255, 0], 2) 364 | cv2.putText(image, 365 | "Bus Count = {}".format(bus_cnt), 366 | (10 , 190), cv2.FONT_HERSHEY_SIMPLEX, 0.5, 367 | [0, 255, 0], 2) 368 | # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>. 369 | cv2.line(image , (340 , 366) , (340 , 138) , (255 , 0 , 0) , 2 ) 370 | #print(1/(time.time()-prev_time)) 371 | out.write(image) 372 | #cv2.imshow('original' , frame_read) 373 | cv2.imshow('Demo', image) 374 | #cv2.imshow('ROI' , ROI) 375 | k = cv2.waitKey(3) 376 | if k&0xFF == ord('q'): 377 | break 378 | else: 379 | print("DONE") 380 | break 381 | cap.release() 382 | out.release() 383 | 384 | if __name__ == "__main__": 385 | YOLO() 386 | -------------------------------------------------------------------------------- /Code/extras/coco.data: -------------------------------------------------------------------------------- 1 | classes= 80 2 | train = /home/pjreddie/data/coco/trainvalno5k.txt 3 | valid = coco_testdev 4 | #valid = data/coco_val_5k.list 5 | names = data/coco.names 6 | backup = /home/pjreddie/backup/ 7 | eval=coco 8 | 9 | -------------------------------------------------------------------------------- /Code/extras/coco.names: -------------------------------------------------------------------------------- 1 | person 2 | bicycle 3 | car 4 | motorbike 5 | aeroplane 6 | bus 7 | train 8 | truck 9 | boat 10 | traffic light 11 | fire hydrant 12 | stop sign 13 | parking meter 14 | bench 15 | bird 16 | cat 17 | dog 18 | horse 19 | sheep 20 | cow 21 | elephant 22 | bear 23 | zebra 24 | giraffe 25 | backpack 26 | umbrella 27 | handbag 28 | tie 29 | suitcase 30 | frisbee 31 | skis 32 | snowboard 33 | sports ball 34 | kite 35 | baseball bat 36 | baseball glove 37 | skateboard 38 | surfboard 39 | tennis racket 40 | bottle 41 | wine glass 42 | cup 43 | fork 44 | knife 45 | spoon 46 | bowl 47 | banana 48 | apple 49 | sandwich 50 | orange 51 | broccoli 52 | carrot 53 | hot dog 54 | pizza 55 | donut 56 | cake 57 | chair 58 | sofa 59 | pottedplant 60 | bed 61 | diningtable 62 | toilet 63 | tvmonitor 64 | laptop 65 | mouse 66 | remote 67 | keyboard 68 | cell phone 69 | microwave 70 | oven 71 | toaster 72 | sink 73 | refrigerator 74 | book 75 | clock 76 | vase 77 | scissors 78 | teddy bear 79 | hair drier 80 | toothbrush 81 | -------------------------------------------------------------------------------- /Code/extras/multi_classify.cfg: -------------------------------------------------------------------------------- 1 | [net] 2 | # Testing 3 | batch=1 4 | subdivisions=1 5 | # Training 6 | #batch=64 7 | #subdivisions=16 8 | width=608 9 | height=608 10 | channels=3 11 | momentum=0.949 12 | decay=0.0005 13 | angle=0 14 | saturation = 1.5 15 | exposure = 1.5 16 | hue=.1 17 | 18 | learning_rate=0.001 19 | burn_in=1000 20 | max_batches = 10000 21 | policy=steps 22 | steps=400000,450000 23 | scales=.1,.1 24 | 25 | #cutmix=1 26 | mosaic=1 27 | 28 | #:104x104 54:52x52 85:26x26 104:13x13 for 416 29 | 30 | [convolutional] 31 | batch_normalize=1 32 | filters=32 33 | size=3 34 | stride=1 35 | pad=1 36 | activation=mish 37 | 38 | # Downsample 39 | 40 | [convolutional] 41 | batch_normalize=1 42 | filters=64 43 | size=3 44 | stride=2 45 | pad=1 46 | activation=mish 47 | 48 | [convolutional] 49 | batch_normalize=1 50 | filters=64 51 | size=1 52 | stride=1 53 | pad=1 54 | activation=mish 55 | 56 | [route] 57 | layers = -2 58 | 59 | [convolutional] 60 | batch_normalize=1 61 | filters=64 62 | size=1 63 | stride=1 64 | pad=1 65 | activation=mish 66 | 67 | [convolutional] 68 | batch_normalize=1 69 | filters=32 70 | size=1 71 | stride=1 72 | pad=1 73 | activation=mish 74 | 75 | [convolutional] 76 | batch_normalize=1 77 | filters=64 78 | size=3 79 | stride=1 80 | pad=1 81 | activation=mish 82 | 83 | [shortcut] 84 | from=-3 85 | activation=linear 86 | 87 | [convolutional] 88 | batch_normalize=1 89 | filters=64 90 | size=1 91 | stride=1 92 | pad=1 93 | activation=mish 94 | 95 | [route] 96 | layers = -1,-7 97 | 98 | [convolutional] 99 | batch_normalize=1 100 | filters=64 101 | size=1 102 | stride=1 103 | pad=1 104 | activation=mish 105 | 106 | # Downsample 107 | 108 | [convolutional] 109 | batch_normalize=1 110 | filters=128 111 | size=3 112 | stride=2 113 | pad=1 114 | activation=mish 115 | 116 | [convolutional] 117 | batch_normalize=1 118 | filters=64 119 | size=1 120 | stride=1 121 | pad=1 122 | activation=mish 123 | 124 | [route] 125 | layers = -2 126 | 127 | [convolutional] 128 | batch_normalize=1 129 | filters=64 130 | size=1 131 | stride=1 132 | pad=1 133 | activation=mish 134 | 135 | [convolutional] 136 | batch_normalize=1 137 | filters=64 138 | size=1 139 | stride=1 140 | pad=1 141 | activation=mish 142 | 143 | [convolutional] 144 | batch_normalize=1 145 | filters=64 146 | size=3 147 | stride=1 148 | pad=1 149 | activation=mish 150 | 151 | [shortcut] 152 | from=-3 153 | activation=linear 154 | 155 | [convolutional] 156 | batch_normalize=1 157 | filters=64 158 | size=1 159 | stride=1 160 | pad=1 161 | activation=mish 162 | 163 | [convolutional] 164 | batch_normalize=1 165 | filters=64 166 | size=3 167 | stride=1 168 | pad=1 169 | activation=mish 170 | 171 | [shortcut] 172 | from=-3 173 | activation=linear 174 | 175 | [convolutional] 176 | batch_normalize=1 177 | filters=64 178 | size=1 179 | stride=1 180 | pad=1 181 | activation=mish 182 | 183 | [route] 184 | layers = -1,-10 185 | 186 | [convolutional] 187 | batch_normalize=1 188 | filters=128 189 | size=1 190 | stride=1 191 | pad=1 192 | activation=mish 193 | 194 | # Downsample 195 | 196 | [convolutional] 197 | batch_normalize=1 198 | filters=256 199 | size=3 200 | stride=2 201 | pad=1 202 | activation=mish 203 | 204 | [convolutional] 205 | batch_normalize=1 206 | filters=128 207 | size=1 208 | stride=1 209 | pad=1 210 | activation=mish 211 | 212 | [route] 213 | layers = -2 214 | 215 | [convolutional] 216 | batch_normalize=1 217 | filters=128 218 | size=1 219 | stride=1 220 | pad=1 221 | activation=mish 222 | 223 | [convolutional] 224 | batch_normalize=1 225 | filters=128 226 | size=1 227 | stride=1 228 | pad=1 229 | activation=mish 230 | 231 | [convolutional] 232 | batch_normalize=1 233 | filters=128 234 | size=3 235 | stride=1 236 | pad=1 237 | activation=mish 238 | 239 | [shortcut] 240 | from=-3 241 | activation=linear 242 | 243 | [convolutional] 244 | batch_normalize=1 245 | filters=128 246 | size=1 247 | stride=1 248 | pad=1 249 | activation=mish 250 | 251 | [convolutional] 252 | batch_normalize=1 253 | filters=128 254 | size=3 255 | stride=1 256 | pad=1 257 | activation=mish 258 | 259 | [shortcut] 260 | from=-3 261 | activation=linear 262 | 263 | [convolutional] 264 | batch_normalize=1 265 | filters=128 266 | size=1 267 | stride=1 268 | pad=1 269 | activation=mish 270 | 271 | [convolutional] 272 | batch_normalize=1 273 | filters=128 274 | size=3 275 | stride=1 276 | pad=1 277 | activation=mish 278 | 279 | [shortcut] 280 | from=-3 281 | activation=linear 282 | 283 | [convolutional] 284 | batch_normalize=1 285 | filters=128 286 | size=1 287 | stride=1 288 | pad=1 289 | activation=mish 290 | 291 | [convolutional] 292 | batch_normalize=1 293 | filters=128 294 | size=3 295 | stride=1 296 | pad=1 297 | activation=mish 298 | 299 | [shortcut] 300 | from=-3 301 | activation=linear 302 | 303 | 304 | [convolutional] 305 | batch_normalize=1 306 | filters=128 307 | size=1 308 | stride=1 309 | pad=1 310 | activation=mish 311 | 312 | [convolutional] 313 | batch_normalize=1 314 | filters=128 315 | size=3 316 | stride=1 317 | pad=1 318 | activation=mish 319 | 320 | [shortcut] 321 | from=-3 322 | activation=linear 323 | 324 | [convolutional] 325 | batch_normalize=1 326 | filters=128 327 | size=1 328 | stride=1 329 | pad=1 330 | activation=mish 331 | 332 | [convolutional] 333 | batch_normalize=1 334 | filters=128 335 | size=3 336 | stride=1 337 | pad=1 338 | activation=mish 339 | 340 | [shortcut] 341 | from=-3 342 | activation=linear 343 | 344 | [convolutional] 345 | batch_normalize=1 346 | filters=128 347 | size=1 348 | stride=1 349 | pad=1 350 | activation=mish 351 | 352 | [convolutional] 353 | batch_normalize=1 354 | filters=128 355 | size=3 356 | stride=1 357 | pad=1 358 | activation=mish 359 | 360 | [shortcut] 361 | from=-3 362 | activation=linear 363 | 364 | [convolutional] 365 | batch_normalize=1 366 | filters=128 367 | size=1 368 | stride=1 369 | pad=1 370 | activation=mish 371 | 372 | [convolutional] 373 | batch_normalize=1 374 | filters=128 375 | size=3 376 | stride=1 377 | pad=1 378 | activation=mish 379 | 380 | [shortcut] 381 | from=-3 382 | activation=linear 383 | 384 | [convolutional] 385 | batch_normalize=1 386 | filters=128 387 | size=1 388 | stride=1 389 | pad=1 390 | activation=mish 391 | 392 | [route] 393 | layers = -1,-28 394 | 395 | [convolutional] 396 | batch_normalize=1 397 | filters=256 398 | size=1 399 | stride=1 400 | pad=1 401 | activation=mish 402 | 403 | # Downsample 404 | 405 | [convolutional] 406 | batch_normalize=1 407 | filters=512 408 | size=3 409 | stride=2 410 | pad=1 411 | activation=mish 412 | 413 | [convolutional] 414 | batch_normalize=1 415 | filters=256 416 | size=1 417 | stride=1 418 | pad=1 419 | activation=mish 420 | 421 | [route] 422 | layers = -2 423 | 424 | [convolutional] 425 | batch_normalize=1 426 | filters=256 427 | size=1 428 | stride=1 429 | pad=1 430 | activation=mish 431 | 432 | [convolutional] 433 | batch_normalize=1 434 | filters=256 435 | size=1 436 | stride=1 437 | pad=1 438 | activation=mish 439 | 440 | [convolutional] 441 | batch_normalize=1 442 | filters=256 443 | size=3 444 | stride=1 445 | pad=1 446 | activation=mish 447 | 448 | [shortcut] 449 | from=-3 450 | activation=linear 451 | 452 | 453 | [convolutional] 454 | batch_normalize=1 455 | filters=256 456 | size=1 457 | stride=1 458 | pad=1 459 | activation=mish 460 | 461 | [convolutional] 462 | batch_normalize=1 463 | filters=256 464 | size=3 465 | stride=1 466 | pad=1 467 | activation=mish 468 | 469 | [shortcut] 470 | from=-3 471 | activation=linear 472 | 473 | 474 | [convolutional] 475 | batch_normalize=1 476 | filters=256 477 | size=1 478 | stride=1 479 | pad=1 480 | activation=mish 481 | 482 | [convolutional] 483 | batch_normalize=1 484 | filters=256 485 | size=3 486 | stride=1 487 | pad=1 488 | activation=mish 489 | 490 | [shortcut] 491 | from=-3 492 | activation=linear 493 | 494 | 495 | [convolutional] 496 | batch_normalize=1 497 | filters=256 498 | size=1 499 | stride=1 500 | pad=1 501 | activation=mish 502 | 503 | [convolutional] 504 | batch_normalize=1 505 | filters=256 506 | size=3 507 | stride=1 508 | pad=1 509 | activation=mish 510 | 511 | [shortcut] 512 | from=-3 513 | activation=linear 514 | 515 | 516 | [convolutional] 517 | batch_normalize=1 518 | filters=256 519 | size=1 520 | stride=1 521 | pad=1 522 | activation=mish 523 | 524 | [convolutional] 525 | batch_normalize=1 526 | filters=256 527 | size=3 528 | stride=1 529 | pad=1 530 | activation=mish 531 | 532 | [shortcut] 533 | from=-3 534 | activation=linear 535 | 536 | 537 | [convolutional] 538 | batch_normalize=1 539 | filters=256 540 | size=1 541 | stride=1 542 | pad=1 543 | activation=mish 544 | 545 | [convolutional] 546 | batch_normalize=1 547 | filters=256 548 | size=3 549 | stride=1 550 | pad=1 551 | activation=mish 552 | 553 | [shortcut] 554 | from=-3 555 | activation=linear 556 | 557 | 558 | [convolutional] 559 | batch_normalize=1 560 | filters=256 561 | size=1 562 | stride=1 563 | pad=1 564 | activation=mish 565 | 566 | [convolutional] 567 | batch_normalize=1 568 | filters=256 569 | size=3 570 | stride=1 571 | pad=1 572 | activation=mish 573 | 574 | [shortcut] 575 | from=-3 576 | activation=linear 577 | 578 | [convolutional] 579 | batch_normalize=1 580 | filters=256 581 | size=1 582 | stride=1 583 | pad=1 584 | activation=mish 585 | 586 | [convolutional] 587 | batch_normalize=1 588 | filters=256 589 | size=3 590 | stride=1 591 | pad=1 592 | activation=mish 593 | 594 | [shortcut] 595 | from=-3 596 | activation=linear 597 | 598 | [convolutional] 599 | batch_normalize=1 600 | filters=256 601 | size=1 602 | stride=1 603 | pad=1 604 | activation=mish 605 | 606 | [route] 607 | layers = -1,-28 608 | 609 | [convolutional] 610 | batch_normalize=1 611 | filters=512 612 | size=1 613 | stride=1 614 | pad=1 615 | activation=mish 616 | 617 | # Downsample 618 | 619 | [convolutional] 620 | batch_normalize=1 621 | filters=1024 622 | size=3 623 | stride=2 624 | pad=1 625 | activation=mish 626 | 627 | [convolutional] 628 | batch_normalize=1 629 | filters=512 630 | size=1 631 | stride=1 632 | pad=1 633 | activation=mish 634 | 635 | [route] 636 | layers = -2 637 | 638 | [convolutional] 639 | batch_normalize=1 640 | filters=512 641 | size=1 642 | stride=1 643 | pad=1 644 | activation=mish 645 | 646 | [convolutional] 647 | batch_normalize=1 648 | filters=512 649 | size=1 650 | stride=1 651 | pad=1 652 | activation=mish 653 | 654 | [convolutional] 655 | batch_normalize=1 656 | filters=512 657 | size=3 658 | stride=1 659 | pad=1 660 | activation=mish 661 | 662 | [shortcut] 663 | from=-3 664 | activation=linear 665 | 666 | [convolutional] 667 | batch_normalize=1 668 | filters=512 669 | size=1 670 | stride=1 671 | pad=1 672 | activation=mish 673 | 674 | [convolutional] 675 | batch_normalize=1 676 | filters=512 677 | size=3 678 | stride=1 679 | pad=1 680 | activation=mish 681 | 682 | [shortcut] 683 | from=-3 684 | activation=linear 685 | 686 | [convolutional] 687 | batch_normalize=1 688 | filters=512 689 | size=1 690 | stride=1 691 | pad=1 692 | activation=mish 693 | 694 | [convolutional] 695 | batch_normalize=1 696 | filters=512 697 | size=3 698 | stride=1 699 | pad=1 700 | activation=mish 701 | 702 | [shortcut] 703 | from=-3 704 | activation=linear 705 | 706 | [convolutional] 707 | batch_normalize=1 708 | filters=512 709 | size=1 710 | stride=1 711 | pad=1 712 | activation=mish 713 | 714 | [convolutional] 715 | batch_normalize=1 716 | filters=512 717 | size=3 718 | stride=1 719 | pad=1 720 | activation=mish 721 | 722 | [shortcut] 723 | from=-3 724 | activation=linear 725 | 726 | [convolutional] 727 | batch_normalize=1 728 | filters=512 729 | size=1 730 | stride=1 731 | pad=1 732 | activation=mish 733 | 734 | [route] 735 | layers = -1,-16 736 | 737 | [convolutional] 738 | batch_normalize=1 739 | filters=1024 740 | size=1 741 | stride=1 742 | pad=1 743 | activation=mish 744 | stopbackward=800 745 | 746 | ########################## 747 | 748 | [convolutional] 749 | batch_normalize=1 750 | filters=512 751 | size=1 752 | stride=1 753 | pad=1 754 | activation=leaky 755 | 756 | [convolutional] 757 | batch_normalize=1 758 | size=3 759 | stride=1 760 | pad=1 761 | filters=1024 762 | activation=leaky 763 | 764 | [convolutional] 765 | batch_normalize=1 766 | filters=512 767 | size=1 768 | stride=1 769 | pad=1 770 | activation=leaky 771 | 772 | ### SPP ### 773 | [maxpool] 774 | stride=1 775 | size=5 776 | 777 | [route] 778 | layers=-2 779 | 780 | [maxpool] 781 | stride=1 782 | size=9 783 | 784 | [route] 785 | layers=-4 786 | 787 | [maxpool] 788 | stride=1 789 | size=13 790 | 791 | [route] 792 | layers=-1,-3,-5,-6 793 | ### End SPP ### 794 | 795 | [convolutional] 796 | batch_normalize=1 797 | filters=512 798 | size=1 799 | stride=1 800 | pad=1 801 | activation=leaky 802 | 803 | [convolutional] 804 | batch_normalize=1 805 | size=3 806 | stride=1 807 | pad=1 808 | filters=1024 809 | activation=leaky 810 | 811 | [convolutional] 812 | batch_normalize=1 813 | filters=512 814 | size=1 815 | stride=1 816 | pad=1 817 | activation=leaky 818 | 819 | [convolutional] 820 | batch_normalize=1 821 | filters=256 822 | size=1 823 | stride=1 824 | pad=1 825 | activation=leaky 826 | 827 | [upsample] 828 | stride=2 829 | 830 | [route] 831 | layers = 85 832 | 833 | [convolutional] 834 | batch_normalize=1 835 | filters=256 836 | size=1 837 | stride=1 838 | pad=1 839 | activation=leaky 840 | 841 | [route] 842 | layers = -1, -3 843 | 844 | [convolutional] 845 | batch_normalize=1 846 | filters=256 847 | size=1 848 | stride=1 849 | pad=1 850 | activation=leaky 851 | 852 | [convolutional] 853 | batch_normalize=1 854 | size=3 855 | stride=1 856 | pad=1 857 | filters=512 858 | activation=leaky 859 | 860 | [convolutional] 861 | batch_normalize=1 862 | filters=256 863 | size=1 864 | stride=1 865 | pad=1 866 | activation=leaky 867 | 868 | [convolutional] 869 | batch_normalize=1 870 | size=3 871 | stride=1 872 | pad=1 873 | filters=512 874 | activation=leaky 875 | 876 | [convolutional] 877 | batch_normalize=1 878 | filters=256 879 | size=1 880 | stride=1 881 | pad=1 882 | activation=leaky 883 | 884 | [convolutional] 885 | batch_normalize=1 886 | filters=128 887 | size=1 888 | stride=1 889 | pad=1 890 | activation=leaky 891 | 892 | [upsample] 893 | stride=2 894 | 895 | [route] 896 | layers = 54 897 | 898 | [convolutional] 899 | batch_normalize=1 900 | filters=128 901 | size=1 902 | stride=1 903 | pad=1 904 | activation=leaky 905 | 906 | [route] 907 | layers = -1, -3 908 | 909 | [convolutional] 910 | batch_normalize=1 911 | filters=128 912 | size=1 913 | stride=1 914 | pad=1 915 | activation=leaky 916 | 917 | [convolutional] 918 | batch_normalize=1 919 | size=3 920 | stride=1 921 | pad=1 922 | filters=256 923 | activation=leaky 924 | 925 | [convolutional] 926 | batch_normalize=1 927 | filters=128 928 | size=1 929 | stride=1 930 | pad=1 931 | activation=leaky 932 | 933 | [convolutional] 934 | batch_normalize=1 935 | size=3 936 | stride=1 937 | pad=1 938 | filters=256 939 | activation=leaky 940 | 941 | [convolutional] 942 | batch_normalize=1 943 | filters=128 944 | size=1 945 | stride=1 946 | pad=1 947 | activation=leaky 948 | 949 | ########################## 950 | 951 | [convolutional] 952 | batch_normalize=1 953 | size=3 954 | stride=1 955 | pad=1 956 | filters=256 957 | activation=leaky 958 | 959 | [convolutional] 960 | size=1 961 | stride=1 962 | pad=1 963 | filters=21 964 | activation=linear 965 | 966 | 967 | [yolo] 968 | mask = 0,1,2 969 | anchors = 12, 16, 19, 36, 40, 28, 36, 75, 76, 55, 72, 146, 142, 110, 192, 243, 459, 401 970 | classes=2 971 | num=9 972 | jitter=.3 973 | ignore_thresh = .7 974 | truth_thresh = 1 975 | scale_x_y = 1.2 976 | iou_thresh=0.213 977 | cls_normalizer=1.0 978 | iou_normalizer=0.07 979 | iou_loss=ciou 980 | nms_kind=greedynms 981 | beta_nms=0.6 982 | max_delta=5 983 | 984 | 985 | [route] 986 | layers = -4 987 | 988 | [convolutional] 989 | batch_normalize=1 990 | size=3 991 | stride=2 992 | pad=1 993 | filters=256 994 | activation=leaky 995 | 996 | [route] 997 | layers = -1, -16 998 | 999 | [convolutional] 1000 | batch_normalize=1 1001 | filters=256 1002 | size=1 1003 | stride=1 1004 | pad=1 1005 | activation=leaky 1006 | 1007 | [convolutional] 1008 | batch_normalize=1 1009 | size=3 1010 | stride=1 1011 | pad=1 1012 | filters=512 1013 | activation=leaky 1014 | 1015 | [convolutional] 1016 | batch_normalize=1 1017 | filters=256 1018 | size=1 1019 | stride=1 1020 | pad=1 1021 | activation=leaky 1022 | 1023 | [convolutional] 1024 | batch_normalize=1 1025 | size=3 1026 | stride=1 1027 | pad=1 1028 | filters=512 1029 | activation=leaky 1030 | 1031 | [convolutional] 1032 | batch_normalize=1 1033 | filters=256 1034 | size=1 1035 | stride=1 1036 | pad=1 1037 | activation=leaky 1038 | 1039 | [convolutional] 1040 | batch_normalize=1 1041 | size=3 1042 | stride=1 1043 | pad=1 1044 | filters=512 1045 | activation=leaky 1046 | 1047 | [convolutional] 1048 | size=1 1049 | stride=1 1050 | pad=1 1051 | filters=21 1052 | activation=linear 1053 | 1054 | 1055 | [yolo] 1056 | mask = 3,4,5 1057 | anchors = 12, 16, 19, 36, 40, 28, 36, 75, 76, 55, 72, 146, 142, 110, 192, 243, 459, 401 1058 | classes=2 1059 | num=9 1060 | jitter=.3 1061 | ignore_thresh = .7 1062 | truth_thresh = 1 1063 | scale_x_y = 1.1 1064 | iou_thresh=0.213 1065 | cls_normalizer=1.0 1066 | iou_normalizer=0.07 1067 | iou_loss=ciou 1068 | nms_kind=greedynms 1069 | beta_nms=0.6 1070 | max_delta=5 1071 | 1072 | 1073 | [route] 1074 | layers = -4 1075 | 1076 | [convolutional] 1077 | batch_normalize=1 1078 | size=3 1079 | stride=2 1080 | pad=1 1081 | filters=512 1082 | activation=leaky 1083 | 1084 | [route] 1085 | layers = -1, -37 1086 | 1087 | [convolutional] 1088 | batch_normalize=1 1089 | filters=512 1090 | size=1 1091 | stride=1 1092 | pad=1 1093 | activation=leaky 1094 | 1095 | [convolutional] 1096 | batch_normalize=1 1097 | size=3 1098 | stride=1 1099 | pad=1 1100 | filters=1024 1101 | activation=leaky 1102 | 1103 | [convolutional] 1104 | batch_normalize=1 1105 | filters=512 1106 | size=1 1107 | stride=1 1108 | pad=1 1109 | activation=leaky 1110 | 1111 | [convolutional] 1112 | batch_normalize=1 1113 | size=3 1114 | stride=1 1115 | pad=1 1116 | filters=1024 1117 | activation=leaky 1118 | 1119 | [convolutional] 1120 | batch_normalize=1 1121 | filters=512 1122 | size=1 1123 | stride=1 1124 | pad=1 1125 | activation=leaky 1126 | 1127 | [convolutional] 1128 | batch_normalize=1 1129 | size=3 1130 | stride=1 1131 | pad=1 1132 | filters=1024 1133 | activation=leaky 1134 | 1135 | [convolutional] 1136 | size=1 1137 | stride=1 1138 | pad=1 1139 | filters=21 1140 | activation=linear 1141 | 1142 | 1143 | [yolo] 1144 | mask = 6,7,8 1145 | anchors = 12, 16, 19, 36, 40, 28, 36, 75, 76, 55, 72, 146, 142, 110, 192, 243, 459, 401 1146 | classes=2 1147 | num=9 1148 | jitter=.3 1149 | ignore_thresh = .7 1150 | truth_thresh = 1 1151 | random=1 1152 | scale_x_y = 1.05 1153 | iou_thresh=0.213 1154 | cls_normalizer=1.0 1155 | iou_normalizer=0.07 1156 | iou_loss=ciou 1157 | nms_kind=greedynms 1158 | beta_nms=0.6 1159 | max_delta=5 1160 | 1161 | -------------------------------------------------------------------------------- /Code/extras/multi_classify.data: -------------------------------------------------------------------------------- 1 | classes= 2 2 | train = data/train.txt 3 | valid = data/test.txt 4 | names = /darknet/data/multi_classify.names 5 | backup = backup/ 6 | -------------------------------------------------------------------------------- /Code/extras/multi_classify.names: -------------------------------------------------------------------------------- 1 | car 2 | wheel 3 | -------------------------------------------------------------------------------- /Code/extras/yolov4.cfg: -------------------------------------------------------------------------------- 1 | [net] 2 | batch=64 3 | subdivisions=8 4 | # Training 5 | #width=416 #512 6 | #height=416 #512 7 | width=608 8 | height=608 9 | channels=3 10 | momentum=0.949 11 | decay=0.0005 12 | angle=0 13 | saturation = 1.5 14 | exposure = 1.5 15 | hue=.1 16 | 17 | learning_rate=0.0013 18 | burn_in=1000 19 | max_batches = 500500 20 | policy=steps 21 | steps=400000,450000 22 | scales=.1,.1 23 | 24 | #cutmix=1 25 | mosaic=1 26 | 27 | #:104x104 54:52x52 85:26x26 104:13x13 for 416 28 | 29 | [convolutional] 30 | batch_normalize=1 31 | filters=32 32 | size=3 33 | stride=1 34 | pad=1 35 | activation=mish 36 | 37 | # Downsample 38 | 39 | [convolutional] 40 | batch_normalize=1 41 | filters=64 42 | size=3 43 | stride=2 44 | pad=1 45 | activation=mish 46 | 47 | [convolutional] 48 | batch_normalize=1 49 | filters=64 50 | size=1 51 | stride=1 52 | pad=1 53 | activation=mish 54 | 55 | [route] 56 | layers = -2 57 | 58 | [convolutional] 59 | batch_normalize=1 60 | filters=64 61 | size=1 62 | stride=1 63 | pad=1 64 | activation=mish 65 | 66 | [convolutional] 67 | batch_normalize=1 68 | filters=32 69 | size=1 70 | stride=1 71 | pad=1 72 | activation=mish 73 | 74 | [convolutional] 75 | batch_normalize=1 76 | filters=64 77 | size=3 78 | stride=1 79 | pad=1 80 | activation=mish 81 | 82 | [shortcut] 83 | from=-3 84 | activation=linear 85 | 86 | [convolutional] 87 | batch_normalize=1 88 | filters=64 89 | size=1 90 | stride=1 91 | pad=1 92 | activation=mish 93 | 94 | [route] 95 | layers = -1,-7 96 | 97 | [convolutional] 98 | batch_normalize=1 99 | filters=64 100 | size=1 101 | stride=1 102 | pad=1 103 | activation=mish 104 | 105 | # Downsample 106 | 107 | [convolutional] 108 | batch_normalize=1 109 | filters=128 110 | size=3 111 | stride=2 112 | pad=1 113 | activation=mish 114 | 115 | [convolutional] 116 | batch_normalize=1 117 | filters=64 118 | size=1 119 | stride=1 120 | pad=1 121 | activation=mish 122 | 123 | [route] 124 | layers = -2 125 | 126 | [convolutional] 127 | batch_normalize=1 128 | filters=64 129 | size=1 130 | stride=1 131 | pad=1 132 | activation=mish 133 | 134 | [convolutional] 135 | batch_normalize=1 136 | filters=64 137 | size=1 138 | stride=1 139 | pad=1 140 | activation=mish 141 | 142 | [convolutional] 143 | batch_normalize=1 144 | filters=64 145 | size=3 146 | stride=1 147 | pad=1 148 | activation=mish 149 | 150 | [shortcut] 151 | from=-3 152 | activation=linear 153 | 154 | [convolutional] 155 | batch_normalize=1 156 | filters=64 157 | size=1 158 | stride=1 159 | pad=1 160 | activation=mish 161 | 162 | [convolutional] 163 | batch_normalize=1 164 | filters=64 165 | size=3 166 | stride=1 167 | pad=1 168 | activation=mish 169 | 170 | [shortcut] 171 | from=-3 172 | activation=linear 173 | 174 | [convolutional] 175 | batch_normalize=1 176 | filters=64 177 | size=1 178 | stride=1 179 | pad=1 180 | activation=mish 181 | 182 | [route] 183 | layers = -1,-10 184 | 185 | [convolutional] 186 | batch_normalize=1 187 | filters=128 188 | size=1 189 | stride=1 190 | pad=1 191 | activation=mish 192 | 193 | # Downsample 194 | 195 | [convolutional] 196 | batch_normalize=1 197 | filters=256 198 | size=3 199 | stride=2 200 | pad=1 201 | activation=mish 202 | 203 | [convolutional] 204 | batch_normalize=1 205 | filters=128 206 | size=1 207 | stride=1 208 | pad=1 209 | activation=mish 210 | 211 | [route] 212 | layers = -2 213 | 214 | [convolutional] 215 | batch_normalize=1 216 | filters=128 217 | size=1 218 | stride=1 219 | pad=1 220 | activation=mish 221 | 222 | [convolutional] 223 | batch_normalize=1 224 | filters=128 225 | size=1 226 | stride=1 227 | pad=1 228 | activation=mish 229 | 230 | [convolutional] 231 | batch_normalize=1 232 | filters=128 233 | size=3 234 | stride=1 235 | pad=1 236 | activation=mish 237 | 238 | [shortcut] 239 | from=-3 240 | activation=linear 241 | 242 | [convolutional] 243 | batch_normalize=1 244 | filters=128 245 | size=1 246 | stride=1 247 | pad=1 248 | activation=mish 249 | 250 | [convolutional] 251 | batch_normalize=1 252 | filters=128 253 | size=3 254 | stride=1 255 | pad=1 256 | activation=mish 257 | 258 | [shortcut] 259 | from=-3 260 | activation=linear 261 | 262 | [convolutional] 263 | batch_normalize=1 264 | filters=128 265 | size=1 266 | stride=1 267 | pad=1 268 | activation=mish 269 | 270 | [convolutional] 271 | batch_normalize=1 272 | filters=128 273 | size=3 274 | stride=1 275 | pad=1 276 | activation=mish 277 | 278 | [shortcut] 279 | from=-3 280 | activation=linear 281 | 282 | [convolutional] 283 | batch_normalize=1 284 | filters=128 285 | size=1 286 | stride=1 287 | pad=1 288 | activation=mish 289 | 290 | [convolutional] 291 | batch_normalize=1 292 | filters=128 293 | size=3 294 | stride=1 295 | pad=1 296 | activation=mish 297 | 298 | [shortcut] 299 | from=-3 300 | activation=linear 301 | 302 | 303 | [convolutional] 304 | batch_normalize=1 305 | filters=128 306 | size=1 307 | stride=1 308 | pad=1 309 | activation=mish 310 | 311 | [convolutional] 312 | batch_normalize=1 313 | filters=128 314 | size=3 315 | stride=1 316 | pad=1 317 | activation=mish 318 | 319 | [shortcut] 320 | from=-3 321 | activation=linear 322 | 323 | [convolutional] 324 | batch_normalize=1 325 | filters=128 326 | size=1 327 | stride=1 328 | pad=1 329 | activation=mish 330 | 331 | [convolutional] 332 | batch_normalize=1 333 | filters=128 334 | size=3 335 | stride=1 336 | pad=1 337 | activation=mish 338 | 339 | [shortcut] 340 | from=-3 341 | activation=linear 342 | 343 | [convolutional] 344 | batch_normalize=1 345 | filters=128 346 | size=1 347 | stride=1 348 | pad=1 349 | activation=mish 350 | 351 | [convolutional] 352 | batch_normalize=1 353 | filters=128 354 | size=3 355 | stride=1 356 | pad=1 357 | activation=mish 358 | 359 | [shortcut] 360 | from=-3 361 | activation=linear 362 | 363 | [convolutional] 364 | batch_normalize=1 365 | filters=128 366 | size=1 367 | stride=1 368 | pad=1 369 | activation=mish 370 | 371 | [convolutional] 372 | batch_normalize=1 373 | filters=128 374 | size=3 375 | stride=1 376 | pad=1 377 | activation=mish 378 | 379 | [shortcut] 380 | from=-3 381 | activation=linear 382 | 383 | [convolutional] 384 | batch_normalize=1 385 | filters=128 386 | size=1 387 | stride=1 388 | pad=1 389 | activation=mish 390 | 391 | [route] 392 | layers = -1,-28 393 | 394 | [convolutional] 395 | batch_normalize=1 396 | filters=256 397 | size=1 398 | stride=1 399 | pad=1 400 | activation=mish 401 | 402 | # Downsample 403 | 404 | [convolutional] 405 | batch_normalize=1 406 | filters=512 407 | size=3 408 | stride=2 409 | pad=1 410 | activation=mish 411 | 412 | [convolutional] 413 | batch_normalize=1 414 | filters=256 415 | size=1 416 | stride=1 417 | pad=1 418 | activation=mish 419 | 420 | [route] 421 | layers = -2 422 | 423 | [convolutional] 424 | batch_normalize=1 425 | filters=256 426 | size=1 427 | stride=1 428 | pad=1 429 | activation=mish 430 | 431 | [convolutional] 432 | batch_normalize=1 433 | filters=256 434 | size=1 435 | stride=1 436 | pad=1 437 | activation=mish 438 | 439 | [convolutional] 440 | batch_normalize=1 441 | filters=256 442 | size=3 443 | stride=1 444 | pad=1 445 | activation=mish 446 | 447 | [shortcut] 448 | from=-3 449 | activation=linear 450 | 451 | 452 | [convolutional] 453 | batch_normalize=1 454 | filters=256 455 | size=1 456 | stride=1 457 | pad=1 458 | activation=mish 459 | 460 | [convolutional] 461 | batch_normalize=1 462 | filters=256 463 | size=3 464 | stride=1 465 | pad=1 466 | activation=mish 467 | 468 | [shortcut] 469 | from=-3 470 | activation=linear 471 | 472 | 473 | [convolutional] 474 | batch_normalize=1 475 | filters=256 476 | size=1 477 | stride=1 478 | pad=1 479 | activation=mish 480 | 481 | [convolutional] 482 | batch_normalize=1 483 | filters=256 484 | size=3 485 | stride=1 486 | pad=1 487 | activation=mish 488 | 489 | [shortcut] 490 | from=-3 491 | activation=linear 492 | 493 | 494 | [convolutional] 495 | batch_normalize=1 496 | filters=256 497 | size=1 498 | stride=1 499 | pad=1 500 | activation=mish 501 | 502 | [convolutional] 503 | batch_normalize=1 504 | filters=256 505 | size=3 506 | stride=1 507 | pad=1 508 | activation=mish 509 | 510 | [shortcut] 511 | from=-3 512 | activation=linear 513 | 514 | 515 | [convolutional] 516 | batch_normalize=1 517 | filters=256 518 | size=1 519 | stride=1 520 | pad=1 521 | activation=mish 522 | 523 | [convolutional] 524 | batch_normalize=1 525 | filters=256 526 | size=3 527 | stride=1 528 | pad=1 529 | activation=mish 530 | 531 | [shortcut] 532 | from=-3 533 | activation=linear 534 | 535 | 536 | [convolutional] 537 | batch_normalize=1 538 | filters=256 539 | size=1 540 | stride=1 541 | pad=1 542 | activation=mish 543 | 544 | [convolutional] 545 | batch_normalize=1 546 | filters=256 547 | size=3 548 | stride=1 549 | pad=1 550 | activation=mish 551 | 552 | [shortcut] 553 | from=-3 554 | activation=linear 555 | 556 | 557 | [convolutional] 558 | batch_normalize=1 559 | filters=256 560 | size=1 561 | stride=1 562 | pad=1 563 | activation=mish 564 | 565 | [convolutional] 566 | batch_normalize=1 567 | filters=256 568 | size=3 569 | stride=1 570 | pad=1 571 | activation=mish 572 | 573 | [shortcut] 574 | from=-3 575 | activation=linear 576 | 577 | [convolutional] 578 | batch_normalize=1 579 | filters=256 580 | size=1 581 | stride=1 582 | pad=1 583 | activation=mish 584 | 585 | [convolutional] 586 | batch_normalize=1 587 | filters=256 588 | size=3 589 | stride=1 590 | pad=1 591 | activation=mish 592 | 593 | [shortcut] 594 | from=-3 595 | activation=linear 596 | 597 | [convolutional] 598 | batch_normalize=1 599 | filters=256 600 | size=1 601 | stride=1 602 | pad=1 603 | activation=mish 604 | 605 | [route] 606 | layers = -1,-28 607 | 608 | [convolutional] 609 | batch_normalize=1 610 | filters=512 611 | size=1 612 | stride=1 613 | pad=1 614 | activation=mish 615 | 616 | # Downsample 617 | 618 | [convolutional] 619 | batch_normalize=1 620 | filters=1024 621 | size=3 622 | stride=2 623 | pad=1 624 | activation=mish 625 | 626 | [convolutional] 627 | batch_normalize=1 628 | filters=512 629 | size=1 630 | stride=1 631 | pad=1 632 | activation=mish 633 | 634 | [route] 635 | layers = -2 636 | 637 | [convolutional] 638 | batch_normalize=1 639 | filters=512 640 | size=1 641 | stride=1 642 | pad=1 643 | activation=mish 644 | 645 | [convolutional] 646 | batch_normalize=1 647 | filters=512 648 | size=1 649 | stride=1 650 | pad=1 651 | activation=mish 652 | 653 | [convolutional] 654 | batch_normalize=1 655 | filters=512 656 | size=3 657 | stride=1 658 | pad=1 659 | activation=mish 660 | 661 | [shortcut] 662 | from=-3 663 | activation=linear 664 | 665 | [convolutional] 666 | batch_normalize=1 667 | filters=512 668 | size=1 669 | stride=1 670 | pad=1 671 | activation=mish 672 | 673 | [convolutional] 674 | batch_normalize=1 675 | filters=512 676 | size=3 677 | stride=1 678 | pad=1 679 | activation=mish 680 | 681 | [shortcut] 682 | from=-3 683 | activation=linear 684 | 685 | [convolutional] 686 | batch_normalize=1 687 | filters=512 688 | size=1 689 | stride=1 690 | pad=1 691 | activation=mish 692 | 693 | [convolutional] 694 | batch_normalize=1 695 | filters=512 696 | size=3 697 | stride=1 698 | pad=1 699 | activation=mish 700 | 701 | [shortcut] 702 | from=-3 703 | activation=linear 704 | 705 | [convolutional] 706 | batch_normalize=1 707 | filters=512 708 | size=1 709 | stride=1 710 | pad=1 711 | activation=mish 712 | 713 | [convolutional] 714 | batch_normalize=1 715 | filters=512 716 | size=3 717 | stride=1 718 | pad=1 719 | activation=mish 720 | 721 | [shortcut] 722 | from=-3 723 | activation=linear 724 | 725 | [convolutional] 726 | batch_normalize=1 727 | filters=512 728 | size=1 729 | stride=1 730 | pad=1 731 | activation=mish 732 | 733 | [route] 734 | layers = -1,-16 735 | 736 | [convolutional] 737 | batch_normalize=1 738 | filters=1024 739 | size=1 740 | stride=1 741 | pad=1 742 | activation=mish 743 | 744 | ########################## 745 | 746 | [convolutional] 747 | batch_normalize=1 748 | filters=512 749 | size=1 750 | stride=1 751 | pad=1 752 | activation=leaky 753 | 754 | [convolutional] 755 | batch_normalize=1 756 | size=3 757 | stride=1 758 | pad=1 759 | filters=1024 760 | activation=leaky 761 | 762 | [convolutional] 763 | batch_normalize=1 764 | filters=512 765 | size=1 766 | stride=1 767 | pad=1 768 | activation=leaky 769 | 770 | ### SPP ### 771 | [maxpool] 772 | stride=1 773 | size=5 774 | 775 | [route] 776 | layers=-2 777 | 778 | [maxpool] 779 | stride=1 780 | size=9 781 | 782 | [route] 783 | layers=-4 784 | 785 | [maxpool] 786 | stride=1 787 | size=13 788 | 789 | [route] 790 | layers=-1,-3,-5,-6 791 | ### End SPP ### 792 | 793 | [convolutional] 794 | batch_normalize=1 795 | filters=512 796 | size=1 797 | stride=1 798 | pad=1 799 | activation=leaky 800 | 801 | [convolutional] 802 | batch_normalize=1 803 | size=3 804 | stride=1 805 | pad=1 806 | filters=1024 807 | activation=leaky 808 | 809 | [convolutional] 810 | batch_normalize=1 811 | filters=512 812 | size=1 813 | stride=1 814 | pad=1 815 | activation=leaky 816 | 817 | [convolutional] 818 | batch_normalize=1 819 | filters=256 820 | size=1 821 | stride=1 822 | pad=1 823 | activation=leaky 824 | 825 | [upsample] 826 | stride=2 827 | 828 | [route] 829 | layers = 85 830 | 831 | [convolutional] 832 | batch_normalize=1 833 | filters=256 834 | size=1 835 | stride=1 836 | pad=1 837 | activation=leaky 838 | 839 | [route] 840 | layers = -1, -3 841 | 842 | [convolutional] 843 | batch_normalize=1 844 | filters=256 845 | size=1 846 | stride=1 847 | pad=1 848 | activation=leaky 849 | 850 | [convolutional] 851 | batch_normalize=1 852 | size=3 853 | stride=1 854 | pad=1 855 | filters=512 856 | activation=leaky 857 | 858 | [convolutional] 859 | batch_normalize=1 860 | filters=256 861 | size=1 862 | stride=1 863 | pad=1 864 | activation=leaky 865 | 866 | [convolutional] 867 | batch_normalize=1 868 | size=3 869 | stride=1 870 | pad=1 871 | filters=512 872 | activation=leaky 873 | 874 | [convolutional] 875 | batch_normalize=1 876 | filters=256 877 | size=1 878 | stride=1 879 | pad=1 880 | activation=leaky 881 | 882 | [convolutional] 883 | batch_normalize=1 884 | filters=128 885 | size=1 886 | stride=1 887 | pad=1 888 | activation=leaky 889 | 890 | [upsample] 891 | stride=2 892 | 893 | [route] 894 | layers = 54 895 | 896 | [convolutional] 897 | batch_normalize=1 898 | filters=128 899 | size=1 900 | stride=1 901 | pad=1 902 | activation=leaky 903 | 904 | [route] 905 | layers = -1, -3 906 | 907 | [convolutional] 908 | batch_normalize=1 909 | filters=128 910 | size=1 911 | stride=1 912 | pad=1 913 | activation=leaky 914 | 915 | [convolutional] 916 | batch_normalize=1 917 | size=3 918 | stride=1 919 | pad=1 920 | filters=256 921 | activation=leaky 922 | 923 | [convolutional] 924 | batch_normalize=1 925 | filters=128 926 | size=1 927 | stride=1 928 | pad=1 929 | activation=leaky 930 | 931 | [convolutional] 932 | batch_normalize=1 933 | size=3 934 | stride=1 935 | pad=1 936 | filters=256 937 | activation=leaky 938 | 939 | [convolutional] 940 | batch_normalize=1 941 | filters=128 942 | size=1 943 | stride=1 944 | pad=1 945 | activation=leaky 946 | 947 | ########################## 948 | 949 | [convolutional] 950 | batch_normalize=1 951 | size=3 952 | stride=1 953 | pad=1 954 | filters=256 955 | activation=leaky 956 | 957 | [convolutional] 958 | size=1 959 | stride=1 960 | pad=1 961 | filters=255 962 | activation=linear 963 | 964 | 965 | [yolo] 966 | mask = 0,1,2 967 | anchors = 12, 16, 19, 36, 40, 28, 36, 75, 76, 55, 72, 146, 142, 110, 192, 243, 459, 401 968 | classes=80 969 | num=9 970 | jitter=.3 971 | ignore_thresh = .7 972 | truth_thresh = 1 973 | scale_x_y = 1.2 974 | iou_thresh=0.213 975 | cls_normalizer=1.0 976 | iou_normalizer=0.07 977 | iou_loss=ciou 978 | nms_kind=greedynms 979 | beta_nms=0.6 980 | max_delta=5 981 | 982 | 983 | [route] 984 | layers = -4 985 | 986 | [convolutional] 987 | batch_normalize=1 988 | size=3 989 | stride=2 990 | pad=1 991 | filters=256 992 | activation=leaky 993 | 994 | [route] 995 | layers = -1, -16 996 | 997 | [convolutional] 998 | batch_normalize=1 999 | filters=256 1000 | size=1 1001 | stride=1 1002 | pad=1 1003 | activation=leaky 1004 | 1005 | [convolutional] 1006 | batch_normalize=1 1007 | size=3 1008 | stride=1 1009 | pad=1 1010 | filters=512 1011 | activation=leaky 1012 | 1013 | [convolutional] 1014 | batch_normalize=1 1015 | filters=256 1016 | size=1 1017 | stride=1 1018 | pad=1 1019 | activation=leaky 1020 | 1021 | [convolutional] 1022 | batch_normalize=1 1023 | size=3 1024 | stride=1 1025 | pad=1 1026 | filters=512 1027 | activation=leaky 1028 | 1029 | [convolutional] 1030 | batch_normalize=1 1031 | filters=256 1032 | size=1 1033 | stride=1 1034 | pad=1 1035 | activation=leaky 1036 | 1037 | [convolutional] 1038 | batch_normalize=1 1039 | size=3 1040 | stride=1 1041 | pad=1 1042 | filters=512 1043 | activation=leaky 1044 | 1045 | [convolutional] 1046 | size=1 1047 | stride=1 1048 | pad=1 1049 | filters=255 1050 | activation=linear 1051 | 1052 | 1053 | [yolo] 1054 | mask = 3,4,5 1055 | anchors = 12, 16, 19, 36, 40, 28, 36, 75, 76, 55, 72, 146, 142, 110, 192, 243, 459, 401 1056 | classes=80 1057 | num=9 1058 | jitter=.3 1059 | ignore_thresh = .7 1060 | truth_thresh = 1 1061 | scale_x_y = 1.1 1062 | iou_thresh=0.213 1063 | cls_normalizer=1.0 1064 | iou_normalizer=0.07 1065 | iou_loss=ciou 1066 | nms_kind=greedynms 1067 | beta_nms=0.6 1068 | max_delta=5 1069 | 1070 | 1071 | [route] 1072 | layers = -4 1073 | 1074 | [convolutional] 1075 | batch_normalize=1 1076 | size=3 1077 | stride=2 1078 | pad=1 1079 | filters=512 1080 | activation=leaky 1081 | 1082 | [route] 1083 | layers = -1, -37 1084 | 1085 | [convolutional] 1086 | batch_normalize=1 1087 | filters=512 1088 | size=1 1089 | stride=1 1090 | pad=1 1091 | activation=leaky 1092 | 1093 | [convolutional] 1094 | batch_normalize=1 1095 | size=3 1096 | stride=1 1097 | pad=1 1098 | filters=1024 1099 | activation=leaky 1100 | 1101 | [convolutional] 1102 | batch_normalize=1 1103 | filters=512 1104 | size=1 1105 | stride=1 1106 | pad=1 1107 | activation=leaky 1108 | 1109 | [convolutional] 1110 | batch_normalize=1 1111 | size=3 1112 | stride=1 1113 | pad=1 1114 | filters=1024 1115 | activation=leaky 1116 | 1117 | [convolutional] 1118 | batch_normalize=1 1119 | filters=512 1120 | size=1 1121 | stride=1 1122 | pad=1 1123 | activation=leaky 1124 | 1125 | [convolutional] 1126 | batch_normalize=1 1127 | size=3 1128 | stride=1 1129 | pad=1 1130 | filters=1024 1131 | activation=leaky 1132 | 1133 | [convolutional] 1134 | size=1 1135 | stride=1 1136 | pad=1 1137 | filters=255 1138 | activation=linear 1139 | 1140 | 1141 | [yolo] 1142 | mask = 6,7,8 1143 | anchors = 12, 16, 19, 36, 40, 28, 36, 75, 76, 55, 72, 146, 142, 110, 192, 243, 459, 401 1144 | classes=80 1145 | num=9 1146 | jitter=.3 1147 | ignore_thresh = .7 1148 | truth_thresh = 1 1149 | random=1 1150 | scale_x_y = 1.05 1151 | iou_thresh=0.213 1152 | cls_normalizer=1.0 1153 | iou_normalizer=0.07 1154 | iou_loss=ciou 1155 | nms_kind=greedynms 1156 | beta_nms=0.6 1157 | max_delta=5 1158 | 1159 | -------------------------------------------------------------------------------- /Code/media/Pics_Readme/GUI.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SravanChittupalli/Advanced-Vehicle-Classifier-and-tracker/d5f53f5eb23b2bba9f23a95941faadf0c1184fab/Code/media/Pics_Readme/GUI.png -------------------------------------------------------------------------------- /Code/media/Pics_Readme/fulldemo.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/SravanChittupalli/Advanced-Vehicle-Classifier-and-tracker/d5f53f5eb23b2bba9f23a95941faadf0c1184fab/Code/media/Pics_Readme/fulldemo.gif -------------------------------------------------------------------------------- /Code/requirements.txt: -------------------------------------------------------------------------------- 1 | numpy==1.19.0 2 | matplotlib==3.0.3 3 | opencv-python==4.2.0.34 4 | opencv-contrib-python==4.2.0.34 5 | Cython==0.29.21 6 | pandas==1.0.5 7 | numba==0.50.1 8 | PyQt5==5.15.0 9 | PyQt5-sip==12.8.0 10 | pyqt5-tools==5.15.0.1.7 11 | scikit-image==0.17.2 12 | scipy==1.5.1 13 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 3, 29 June 2007 3 | 4 | Copyright (C) 2007 Free Software Foundation, Inc. 5 | Everyone is permitted to copy and distribute verbatim copies 6 | of this license document, but changing it is not allowed. 7 | 8 | Preamble 9 | 10 | The GNU General Public License is a free, copyleft license for 11 | software and other kinds of works. 12 | 13 | The licenses for most software and other practical works are designed 14 | to take away your freedom to share and change the works. By contrast, 15 | the GNU General Public License is intended to guarantee your freedom to 16 | share and change all versions of a program--to make sure it remains free 17 | software for all its users. We, the Free Software Foundation, use the 18 | GNU General Public License for most of our software; it applies also to 19 | any other work released this way by its authors. You can apply it to 20 | your programs, too. 21 | 22 | When we speak of free software, we are referring to freedom, not 23 | price. Our General Public Licenses are designed to make sure that you 24 | have the freedom to distribute copies of free software (and charge for 25 | them if you wish), that you receive source code or can get it if you 26 | want it, that you can change the software or use pieces of it in new 27 | free programs, and that you know you can do these things. 28 | 29 | To protect your rights, we need to prevent others from denying you 30 | these rights or asking you to surrender the rights. Therefore, you have 31 | certain responsibilities if you distribute copies of the software, or if 32 | you modify it: responsibilities to respect the freedom of others. 33 | 34 | For example, if you distribute copies of such a program, whether 35 | gratis or for a fee, you must pass on to the recipients the same 36 | freedoms that you received. You must make sure that they, too, receive 37 | or can get the source code. And you must show them these terms so they 38 | know their rights. 39 | 40 | Developers that use the GNU GPL protect your rights with two steps: 41 | (1) assert copyright on the software, and (2) offer you this License 42 | giving you legal permission to copy, distribute and/or modify it. 43 | 44 | For the developers' and authors' protection, the GPL clearly explains 45 | that there is no warranty for this free software. For both users' and 46 | authors' sake, the GPL requires that modified versions be marked as 47 | changed, so that their problems will not be attributed erroneously to 48 | authors of previous versions. 49 | 50 | Some devices are designed to deny users access to install or run 51 | modified versions of the software inside them, although the manufacturer 52 | can do so. This is fundamentally incompatible with the aim of 53 | protecting users' freedom to change the software. The systematic 54 | pattern of such abuse occurs in the area of products for individuals to 55 | use, which is precisely where it is most unacceptable. Therefore, we 56 | have designed this version of the GPL to prohibit the practice for those 57 | products. If such problems arise substantially in other domains, we 58 | stand ready to extend this provision to those domains in future versions 59 | of the GPL, as needed to protect the freedom of users. 60 | 61 | Finally, every program is threatened constantly by software patents. 62 | States should not allow patents to restrict development and use of 63 | software on general-purpose computers, but in those that do, we wish to 64 | avoid the special danger that patents applied to a free program could 65 | make it effectively proprietary. To prevent this, the GPL assures that 66 | patents cannot be used to render the program non-free. 67 | 68 | The precise terms and conditions for copying, distribution and 69 | modification follow. 70 | 71 | TERMS AND CONDITIONS 72 | 73 | 0. Definitions. 74 | 75 | "This License" refers to version 3 of the GNU General Public License. 76 | 77 | "Copyright" also means copyright-like laws that apply to other kinds of 78 | works, such as semiconductor masks. 79 | 80 | "The Program" refers to any copyrightable work licensed under this 81 | License. Each licensee is addressed as "you". "Licensees" and 82 | "recipients" may be individuals or organizations. 83 | 84 | To "modify" a work means to copy from or adapt all or part of the work 85 | in a fashion requiring copyright permission, other than the making of an 86 | exact copy. The resulting work is called a "modified version" of the 87 | earlier work or a work "based on" the earlier work. 88 | 89 | A "covered work" means either the unmodified Program or a work based 90 | on the Program. 91 | 92 | To "propagate" a work means to do anything with it that, without 93 | permission, would make you directly or secondarily liable for 94 | infringement under applicable copyright law, except executing it on a 95 | computer or modifying a private copy. Propagation includes copying, 96 | distribution (with or without modification), making available to the 97 | public, and in some countries other activities as well. 98 | 99 | To "convey" a work means any kind of propagation that enables other 100 | parties to make or receive copies. Mere interaction with a user through 101 | a computer network, with no transfer of a copy, is not conveying. 102 | 103 | An interactive user interface displays "Appropriate Legal Notices" 104 | to the extent that it includes a convenient and prominently visible 105 | feature that (1) displays an appropriate copyright notice, and (2) 106 | tells the user that there is no warranty for the work (except to the 107 | extent that warranties are provided), that licensees may convey the 108 | work under this License, and how to view a copy of this License. If 109 | the interface presents a list of user commands or options, such as a 110 | menu, a prominent item in the list meets this criterion. 111 | 112 | 1. Source Code. 113 | 114 | The "source code" for a work means the preferred form of the work 115 | for making modifications to it. "Object code" means any non-source 116 | form of a work. 117 | 118 | A "Standard Interface" means an interface that either is an official 119 | standard defined by a recognized standards body, or, in the case of 120 | interfaces specified for a particular programming language, one that 121 | is widely used among developers working in that language. 122 | 123 | The "System Libraries" of an executable work include anything, other 124 | than the work as a whole, that (a) is included in the normal form of 125 | packaging a Major Component, but which is not part of that Major 126 | Component, and (b) serves only to enable use of the work with that 127 | Major Component, or to implement a Standard Interface for which an 128 | implementation is available to the public in source code form. A 129 | "Major Component", in this context, means a major essential component 130 | (kernel, window system, and so on) of the specific operating system 131 | (if any) on which the executable work runs, or a compiler used to 132 | produce the work, or an object code interpreter used to run it. 133 | 134 | The "Corresponding Source" for a work in object code form means all 135 | the source code needed to generate, install, and (for an executable 136 | work) run the object code and to modify the work, including scripts to 137 | control those activities. However, it does not include the work's 138 | System Libraries, or general-purpose tools or generally available free 139 | programs which are used unmodified in performing those activities but 140 | which are not part of the work. For example, Corresponding Source 141 | includes interface definition files associated with source files for 142 | the work, and the source code for shared libraries and dynamically 143 | linked subprograms that the work is specifically designed to require, 144 | such as by intimate data communication or control flow between those 145 | subprograms and other parts of the work. 146 | 147 | The Corresponding Source need not include anything that users 148 | can regenerate automatically from other parts of the Corresponding 149 | Source. 150 | 151 | The Corresponding Source for a work in source code form is that 152 | same work. 153 | 154 | 2. Basic Permissions. 155 | 156 | All rights granted under this License are granted for the term of 157 | copyright on the Program, and are irrevocable provided the stated 158 | conditions are met. This License explicitly affirms your unlimited 159 | permission to run the unmodified Program. The output from running a 160 | covered work is covered by this License only if the output, given its 161 | content, constitutes a covered work. This License acknowledges your 162 | rights of fair use or other equivalent, as provided by copyright law. 163 | 164 | You may make, run and propagate covered works that you do not 165 | convey, without conditions so long as your license otherwise remains 166 | in force. You may convey covered works to others for the sole purpose 167 | of having them make modifications exclusively for you, or provide you 168 | with facilities for running those works, provided that you comply with 169 | the terms of this License in conveying all material for which you do 170 | not control copyright. Those thus making or running the covered works 171 | for you must do so exclusively on your behalf, under your direction 172 | and control, on terms that prohibit them from making any copies of 173 | your copyrighted material outside their relationship with you. 174 | 175 | Conveying under any other circumstances is permitted solely under 176 | the conditions stated below. Sublicensing is not allowed; section 10 177 | makes it unnecessary. 178 | 179 | 3. Protecting Users' Legal Rights From Anti-Circumvention Law. 180 | 181 | No covered work shall be deemed part of an effective technological 182 | measure under any applicable law fulfilling obligations under article 183 | 11 of the WIPO copyright treaty adopted on 20 December 1996, or 184 | similar laws prohibiting or restricting circumvention of such 185 | measures. 186 | 187 | When you convey a covered work, you waive any legal power to forbid 188 | circumvention of technological measures to the extent such circumvention 189 | is effected by exercising rights under this License with respect to 190 | the covered work, and you disclaim any intention to limit operation or 191 | modification of the work as a means of enforcing, against the work's 192 | users, your or third parties' legal rights to forbid circumvention of 193 | technological measures. 194 | 195 | 4. Conveying Verbatim Copies. 196 | 197 | You may convey verbatim copies of the Program's source code as you 198 | receive it, in any medium, provided that you conspicuously and 199 | appropriately publish on each copy an appropriate copyright notice; 200 | keep intact all notices stating that this License and any 201 | non-permissive terms added in accord with section 7 apply to the code; 202 | keep intact all notices of the absence of any warranty; and give all 203 | recipients a copy of this License along with the Program. 204 | 205 | You may charge any price or no price for each copy that you convey, 206 | and you may offer support or warranty protection for a fee. 207 | 208 | 5. Conveying Modified Source Versions. 209 | 210 | You may convey a work based on the Program, or the modifications to 211 | produce it from the Program, in the form of source code under the 212 | terms of section 4, provided that you also meet all of these conditions: 213 | 214 | a) The work must carry prominent notices stating that you modified 215 | it, and giving a relevant date. 216 | 217 | b) The work must carry prominent notices stating that it is 218 | released under this License and any conditions added under section 219 | 7. This requirement modifies the requirement in section 4 to 220 | "keep intact all notices". 221 | 222 | c) You must license the entire work, as a whole, under this 223 | License to anyone who comes into possession of a copy. This 224 | License will therefore apply, along with any applicable section 7 225 | additional terms, to the whole of the work, and all its parts, 226 | regardless of how they are packaged. This License gives no 227 | permission to license the work in any other way, but it does not 228 | invalidate such permission if you have separately received it. 229 | 230 | d) If the work has interactive user interfaces, each must display 231 | Appropriate Legal Notices; however, if the Program has interactive 232 | interfaces that do not display Appropriate Legal Notices, your 233 | work need not make them do so. 234 | 235 | A compilation of a covered work with other separate and independent 236 | works, which are not by their nature extensions of the covered work, 237 | and which are not combined with it such as to form a larger program, 238 | in or on a volume of a storage or distribution medium, is called an 239 | "aggregate" if the compilation and its resulting copyright are not 240 | used to limit the access or legal rights of the compilation's users 241 | beyond what the individual works permit. Inclusion of a covered work 242 | in an aggregate does not cause this License to apply to the other 243 | parts of the aggregate. 244 | 245 | 6. Conveying Non-Source Forms. 246 | 247 | You may convey a covered work in object code form under the terms 248 | of sections 4 and 5, provided that you also convey the 249 | machine-readable Corresponding Source under the terms of this License, 250 | in one of these ways: 251 | 252 | a) Convey the object code in, or embodied in, a physical product 253 | (including a physical distribution medium), accompanied by the 254 | Corresponding Source fixed on a durable physical medium 255 | customarily used for software interchange. 256 | 257 | b) Convey the object code in, or embodied in, a physical product 258 | (including a physical distribution medium), accompanied by a 259 | written offer, valid for at least three years and valid for as 260 | long as you offer spare parts or customer support for that product 261 | model, to give anyone who possesses the object code either (1) a 262 | copy of the Corresponding Source for all the software in the 263 | product that is covered by this License, on a durable physical 264 | medium customarily used for software interchange, for a price no 265 | more than your reasonable cost of physically performing this 266 | conveying of source, or (2) access to copy the 267 | Corresponding Source from a network server at no charge. 268 | 269 | c) Convey individual copies of the object code with a copy of the 270 | written offer to provide the Corresponding Source. This 271 | alternative is allowed only occasionally and noncommercially, and 272 | only if you received the object code with such an offer, in accord 273 | with subsection 6b. 274 | 275 | d) Convey the object code by offering access from a designated 276 | place (gratis or for a charge), and offer equivalent access to the 277 | Corresponding Source in the same way through the same place at no 278 | further charge. You need not require recipients to copy the 279 | Corresponding Source along with the object code. If the place to 280 | copy the object code is a network server, the Corresponding Source 281 | may be on a different server (operated by you or a third party) 282 | that supports equivalent copying facilities, provided you maintain 283 | clear directions next to the object code saying where to find the 284 | Corresponding Source. Regardless of what server hosts the 285 | Corresponding Source, you remain obligated to ensure that it is 286 | available for as long as needed to satisfy these requirements. 287 | 288 | e) Convey the object code using peer-to-peer transmission, provided 289 | you inform other peers where the object code and Corresponding 290 | Source of the work are being offered to the general public at no 291 | charge under subsection 6d. 292 | 293 | A separable portion of the object code, whose source code is excluded 294 | from the Corresponding Source as a System Library, need not be 295 | included in conveying the object code work. 296 | 297 | A "User Product" is either (1) a "consumer product", which means any 298 | tangible personal property which is normally used for personal, family, 299 | or household purposes, or (2) anything designed or sold for incorporation 300 | into a dwelling. In determining whether a product is a consumer product, 301 | doubtful cases shall be resolved in favor of coverage. For a particular 302 | product received by a particular user, "normally used" refers to a 303 | typical or common use of that class of product, regardless of the status 304 | of the particular user or of the way in which the particular user 305 | actually uses, or expects or is expected to use, the product. A product 306 | is a consumer product regardless of whether the product has substantial 307 | commercial, industrial or non-consumer uses, unless such uses represent 308 | the only significant mode of use of the product. 309 | 310 | "Installation Information" for a User Product means any methods, 311 | procedures, authorization keys, or other information required to install 312 | and execute modified versions of a covered work in that User Product from 313 | a modified version of its Corresponding Source. The information must 314 | suffice to ensure that the continued functioning of the modified object 315 | code is in no case prevented or interfered with solely because 316 | modification has been made. 317 | 318 | If you convey an object code work under this section in, or with, or 319 | specifically for use in, a User Product, and the conveying occurs as 320 | part of a transaction in which the right of possession and use of the 321 | User Product is transferred to the recipient in perpetuity or for a 322 | fixed term (regardless of how the transaction is characterized), the 323 | Corresponding Source conveyed under this section must be accompanied 324 | by the Installation Information. But this requirement does not apply 325 | if neither you nor any third party retains the ability to install 326 | modified object code on the User Product (for example, the work has 327 | been installed in ROM). 328 | 329 | The requirement to provide Installation Information does not include a 330 | requirement to continue to provide support service, warranty, or updates 331 | for a work that has been modified or installed by the recipient, or for 332 | the User Product in which it has been modified or installed. Access to a 333 | network may be denied when the modification itself materially and 334 | adversely affects the operation of the network or violates the rules and 335 | protocols for communication across the network. 336 | 337 | Corresponding Source conveyed, and Installation Information provided, 338 | in accord with this section must be in a format that is publicly 339 | documented (and with an implementation available to the public in 340 | source code form), and must require no special password or key for 341 | unpacking, reading or copying. 342 | 343 | 7. Additional Terms. 344 | 345 | "Additional permissions" are terms that supplement the terms of this 346 | License by making exceptions from one or more of its conditions. 347 | Additional permissions that are applicable to the entire Program shall 348 | be treated as though they were included in this License, to the extent 349 | that they are valid under applicable law. If additional permissions 350 | apply only to part of the Program, that part may be used separately 351 | under those permissions, but the entire Program remains governed by 352 | this License without regard to the additional permissions. 353 | 354 | When you convey a copy of a covered work, you may at your option 355 | remove any additional permissions from that copy, or from any part of 356 | it. (Additional permissions may be written to require their own 357 | removal in certain cases when you modify the work.) You may place 358 | additional permissions on material, added by you to a covered work, 359 | for which you have or can give appropriate copyright permission. 360 | 361 | Notwithstanding any other provision of this License, for material you 362 | add to a covered work, you may (if authorized by the copyright holders of 363 | that material) supplement the terms of this License with terms: 364 | 365 | a) Disclaiming warranty or limiting liability differently from the 366 | terms of sections 15 and 16 of this License; or 367 | 368 | b) Requiring preservation of specified reasonable legal notices or 369 | author attributions in that material or in the Appropriate Legal 370 | Notices displayed by works containing it; or 371 | 372 | c) Prohibiting misrepresentation of the origin of that material, or 373 | requiring that modified versions of such material be marked in 374 | reasonable ways as different from the original version; or 375 | 376 | d) Limiting the use for publicity purposes of names of licensors or 377 | authors of the material; or 378 | 379 | e) Declining to grant rights under trademark law for use of some 380 | trade names, trademarks, or service marks; or 381 | 382 | f) Requiring indemnification of licensors and authors of that 383 | material by anyone who conveys the material (or modified versions of 384 | it) with contractual assumptions of liability to the recipient, for 385 | any liability that these contractual assumptions directly impose on 386 | those licensors and authors. 387 | 388 | All other non-permissive additional terms are considered "further 389 | restrictions" within the meaning of section 10. If the Program as you 390 | received it, or any part of it, contains a notice stating that it is 391 | governed by this License along with a term that is a further 392 | restriction, you may remove that term. If a license document contains 393 | a further restriction but permits relicensing or conveying under this 394 | License, you may add to a covered work material governed by the terms 395 | of that license document, provided that the further restriction does 396 | not survive such relicensing or conveying. 397 | 398 | If you add terms to a covered work in accord with this section, you 399 | must place, in the relevant source files, a statement of the 400 | additional terms that apply to those files, or a notice indicating 401 | where to find the applicable terms. 402 | 403 | Additional terms, permissive or non-permissive, may be stated in the 404 | form of a separately written license, or stated as exceptions; 405 | the above requirements apply either way. 406 | 407 | 8. Termination. 408 | 409 | You may not propagate or modify a covered work except as expressly 410 | provided under this License. Any attempt otherwise to propagate or 411 | modify it is void, and will automatically terminate your rights under 412 | this License (including any patent licenses granted under the third 413 | paragraph of section 11). 414 | 415 | However, if you cease all violation of this License, then your 416 | license from a particular copyright holder is reinstated (a) 417 | provisionally, unless and until the copyright holder explicitly and 418 | finally terminates your license, and (b) permanently, if the copyright 419 | holder fails to notify you of the violation by some reasonable means 420 | prior to 60 days after the cessation. 421 | 422 | Moreover, your license from a particular copyright holder is 423 | reinstated permanently if the copyright holder notifies you of the 424 | violation by some reasonable means, this is the first time you have 425 | received notice of violation of this License (for any work) from that 426 | copyright holder, and you cure the violation prior to 30 days after 427 | your receipt of the notice. 428 | 429 | Termination of your rights under this section does not terminate the 430 | licenses of parties who have received copies or rights from you under 431 | this License. If your rights have been terminated and not permanently 432 | reinstated, you do not qualify to receive new licenses for the same 433 | material under section 10. 434 | 435 | 9. Acceptance Not Required for Having Copies. 436 | 437 | You are not required to accept this License in order to receive or 438 | run a copy of the Program. Ancillary propagation of a covered work 439 | occurring solely as a consequence of using peer-to-peer transmission 440 | to receive a copy likewise does not require acceptance. However, 441 | nothing other than this License grants you permission to propagate or 442 | modify any covered work. These actions infringe copyright if you do 443 | not accept this License. Therefore, by modifying or propagating a 444 | covered work, you indicate your acceptance of this License to do so. 445 | 446 | 10. Automatic Licensing of Downstream Recipients. 447 | 448 | Each time you convey a covered work, the recipient automatically 449 | receives a license from the original licensors, to run, modify and 450 | propagate that work, subject to this License. You are not responsible 451 | for enforcing compliance by third parties with this License. 452 | 453 | An "entity transaction" is a transaction transferring control of an 454 | organization, or substantially all assets of one, or subdividing an 455 | organization, or merging organizations. If propagation of a covered 456 | work results from an entity transaction, each party to that 457 | transaction who receives a copy of the work also receives whatever 458 | licenses to the work the party's predecessor in interest had or could 459 | give under the previous paragraph, plus a right to possession of the 460 | Corresponding Source of the work from the predecessor in interest, if 461 | the predecessor has it or can get it with reasonable efforts. 462 | 463 | You may not impose any further restrictions on the exercise of the 464 | rights granted or affirmed under this License. For example, you may 465 | not impose a license fee, royalty, or other charge for exercise of 466 | rights granted under this License, and you may not initiate litigation 467 | (including a cross-claim or counterclaim in a lawsuit) alleging that 468 | any patent claim is infringed by making, using, selling, offering for 469 | sale, or importing the Program or any portion of it. 470 | 471 | 11. Patents. 472 | 473 | A "contributor" is a copyright holder who authorizes use under this 474 | License of the Program or a work on which the Program is based. The 475 | work thus licensed is called the contributor's "contributor version". 476 | 477 | A contributor's "essential patent claims" are all patent claims 478 | owned or controlled by the contributor, whether already acquired or 479 | hereafter acquired, that would be infringed by some manner, permitted 480 | by this License, of making, using, or selling its contributor version, 481 | but do not include claims that would be infringed only as a 482 | consequence of further modification of the contributor version. For 483 | purposes of this definition, "control" includes the right to grant 484 | patent sublicenses in a manner consistent with the requirements of 485 | this License. 486 | 487 | Each contributor grants you a non-exclusive, worldwide, royalty-free 488 | patent license under the contributor's essential patent claims, to 489 | make, use, sell, offer for sale, import and otherwise run, modify and 490 | propagate the contents of its contributor version. 491 | 492 | In the following three paragraphs, a "patent license" is any express 493 | agreement or commitment, however denominated, not to enforce a patent 494 | (such as an express permission to practice a patent or covenant not to 495 | sue for patent infringement). To "grant" such a patent license to a 496 | party means to make such an agreement or commitment not to enforce a 497 | patent against the party. 498 | 499 | If you convey a covered work, knowingly relying on a patent license, 500 | and the Corresponding Source of the work is not available for anyone 501 | to copy, free of charge and under the terms of this License, through a 502 | publicly available network server or other readily accessible means, 503 | then you must either (1) cause the Corresponding Source to be so 504 | available, or (2) arrange to deprive yourself of the benefit of the 505 | patent license for this particular work, or (3) arrange, in a manner 506 | consistent with the requirements of this License, to extend the patent 507 | license to downstream recipients. "Knowingly relying" means you have 508 | actual knowledge that, but for the patent license, your conveying the 509 | covered work in a country, or your recipient's use of the covered work 510 | in a country, would infringe one or more identifiable patents in that 511 | country that you have reason to believe are valid. 512 | 513 | If, pursuant to or in connection with a single transaction or 514 | arrangement, you convey, or propagate by procuring conveyance of, a 515 | covered work, and grant a patent license to some of the parties 516 | receiving the covered work authorizing them to use, propagate, modify 517 | or convey a specific copy of the covered work, then the patent license 518 | you grant is automatically extended to all recipients of the covered 519 | work and works based on it. 520 | 521 | A patent license is "discriminatory" if it does not include within 522 | the scope of its coverage, prohibits the exercise of, or is 523 | conditioned on the non-exercise of one or more of the rights that are 524 | specifically granted under this License. You may not convey a covered 525 | work if you are a party to an arrangement with a third party that is 526 | in the business of distributing software, under which you make payment 527 | to the third party based on the extent of your activity of conveying 528 | the work, and under which the third party grants, to any of the 529 | parties who would receive the covered work from you, a discriminatory 530 | patent license (a) in connection with copies of the covered work 531 | conveyed by you (or copies made from those copies), or (b) primarily 532 | for and in connection with specific products or compilations that 533 | contain the covered work, unless you entered into that arrangement, 534 | or that patent license was granted, prior to 28 March 2007. 535 | 536 | Nothing in this License shall be construed as excluding or limiting 537 | any implied license or other defenses to infringement that may 538 | otherwise be available to you under applicable patent law. 539 | 540 | 12. No Surrender of Others' Freedom. 541 | 542 | If conditions are imposed on you (whether by court order, agreement or 543 | otherwise) that contradict the conditions of this License, they do not 544 | excuse you from the conditions of this License. If you cannot convey a 545 | covered work so as to satisfy simultaneously your obligations under this 546 | License and any other pertinent obligations, then as a consequence you may 547 | not convey it at all. For example, if you agree to terms that obligate you 548 | to collect a royalty for further conveying from those to whom you convey 549 | the Program, the only way you could satisfy both those terms and this 550 | License would be to refrain entirely from conveying the Program. 551 | 552 | 13. Use with the GNU Affero General Public License. 553 | 554 | Notwithstanding any other provision of this License, you have 555 | permission to link or combine any covered work with a work licensed 556 | under version 3 of the GNU Affero General Public License into a single 557 | combined work, and to convey the resulting work. The terms of this 558 | License will continue to apply to the part which is the covered work, 559 | but the special requirements of the GNU Affero General Public License, 560 | section 13, concerning interaction through a network will apply to the 561 | combination as such. 562 | 563 | 14. Revised Versions of this License. 564 | 565 | The Free Software Foundation may publish revised and/or new versions of 566 | the GNU General Public License from time to time. Such new versions will 567 | be similar in spirit to the present version, but may differ in detail to 568 | address new problems or concerns. 569 | 570 | Each version is given a distinguishing version number. If the 571 | Program specifies that a certain numbered version of the GNU General 572 | Public License "or any later version" applies to it, you have the 573 | option of following the terms and conditions either of that numbered 574 | version or of any later version published by the Free Software 575 | Foundation. If the Program does not specify a version number of the 576 | GNU General Public License, you may choose any version ever published 577 | by the Free Software Foundation. 578 | 579 | If the Program specifies that a proxy can decide which future 580 | versions of the GNU General Public License can be used, that proxy's 581 | public statement of acceptance of a version permanently authorizes you 582 | to choose that version for the Program. 583 | 584 | Later license versions may give you additional or different 585 | permissions. However, no additional obligations are imposed on any 586 | author or copyright holder as a result of your choosing to follow a 587 | later version. 588 | 589 | 15. Disclaimer of Warranty. 590 | 591 | THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY 592 | APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT 593 | HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY 594 | OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, 595 | THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 596 | PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM 597 | IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF 598 | ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 599 | 600 | 16. Limitation of Liability. 601 | 602 | IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING 603 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS 604 | THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY 605 | GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE 606 | USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF 607 | DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD 608 | PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), 609 | EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF 610 | SUCH DAMAGES. 611 | 612 | 17. Interpretation of Sections 15 and 16. 613 | 614 | If the disclaimer of warranty and limitation of liability provided 615 | above cannot be given local legal effect according to their terms, 616 | reviewing courts shall apply local law that most closely approximates 617 | an absolute waiver of all civil liability in connection with the 618 | Program, unless a warranty or assumption of liability accompanies a 619 | copy of the Program in return for a fee. 620 | 621 | END OF TERMS AND CONDITIONS 622 | 623 | How to Apply These Terms to Your New Programs 624 | 625 | If you develop a new program, and you want it to be of the greatest 626 | possible use to the public, the best way to achieve this is to make it 627 | free software which everyone can redistribute and change under these terms. 628 | 629 | To do so, attach the following notices to the program. It is safest 630 | to attach them to the start of each source file to most effectively 631 | state the exclusion of warranty; and each file should have at least 632 | the "copyright" line and a pointer to where the full notice is found. 633 | 634 | 635 | Copyright (C) 636 | 637 | This program is free software: you can redistribute it and/or modify 638 | it under the terms of the GNU General Public License as published by 639 | the Free Software Foundation, either version 3 of the License, or 640 | (at your option) any later version. 641 | 642 | This program is distributed in the hope that it will be useful, 643 | but WITHOUT ANY WARRANTY; without even the implied warranty of 644 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 645 | GNU General Public License for more details. 646 | 647 | You should have received a copy of the GNU General Public License 648 | along with this program. If not, see . 649 | 650 | Also add information on how to contact you by electronic and paper mail. 651 | 652 | If the program does terminal interaction, make it output a short 653 | notice like this when it starts in an interactive mode: 654 | 655 | Copyright (C) 656 | This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. 657 | This is free software, and you are welcome to redistribute it 658 | under certain conditions; type `show c' for details. 659 | 660 | The hypothetical commands `show w' and `show c' should show the appropriate 661 | parts of the General Public License. Of course, your program's commands 662 | might be different; for a GUI interface, you would use an "about box". 663 | 664 | You should also get your employer (if you work as a programmer) or school, 665 | if any, to sign a "copyright disclaimer" for the program, if necessary. 666 | For more information on this, and how to apply and follow the GNU GPL, see 667 | . 668 | 669 | The GNU General Public License does not permit incorporating your program 670 | into proprietary programs. If your program is a subroutine library, you 671 | may consider it more useful to permit linking proprietary applications with 672 | the library. If this is what you want to do, use the GNU Lesser General 673 | Public License instead of this License. But first, please read 674 | . 675 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Advanced-Vehicle-Classifier-And-Tracker 🚗 🚛 🚴🏽 🚍 2 | This project aims at classifying multiple classes of vehicles and in addition to that sub-classifying of trucks according to number of axels. 3 | 4 | ## Contents 5 | - [Why this project](#why-this-project) 6 | - [About the Project](#about-the-project) 7 | - [Tech Stack](#tech-stack) 8 | - [File Structure](#file-structure) 9 | - [How to run the application](#how-to-run-the-application) 10 | - [Demo](#demo) 11 | - [To-Do](#to-do) 12 | - [References](#references) 13 | - [License](#license) 14 | 15 | ## Why this project 16 | - Sub-classification of trucks is a very important task as it has many uses like toll-automation and counting number and type of vehicles accurately. Specially in India all types of trucks look the same untill you are a truck connoisseur but the wear and tear done by different trucks depends on the number of tyres. So each vehicle has to be charged differently. 17 | 18 | - You can also see that at toll gates where it is completely manual the workers can misinform the authorities about the total amount earned for that day thus leading to loss for the authority owning the road. This project can reduce such frauds. 19 | 20 | - Nowadays we find the use of RFID tags for toll automation. The problem with this is every user has to buy one but by using _ComputerVision_ no external hardware is required. Once setup you will get the data automatically 24X7. 21 | 22 | ## About the Project 23 | ![GUI](https://github.com/SravanChittupalli/Advanced-Vehicle-Classifier/blob/master/Code/media/Pics_Readme/GUI.png) 24 | 25 | The project contains a GUI application to generalise the project. It can be used at toll gates to keep an accurate track of number of vehicles and with addition of some more classes and features the project can also be used for full toll automation. The project uses Darknet YOLOv4 and SORT tracker. For now the app can only classify car , bike , bus and 5 classes of trucks. 26 | 27 | ## Tech Stack 28 | * [Python](https://www.python.org/) 29 | * [Numpy](https://numpy.org) 30 | * [SciPy](https://pypi.org/project/scipy/1.5.1/) 31 | * [OpenCV](https://opencv.org/) 32 | * [PyQt5](https://pypi.org/project/PyQt5/) 33 | 34 | 35 | ## File Structure 36 | ├── LICENSE 37 | ├── README.md 38 | ├── WORKPLAN.md # Project plan 39 | ├── Code/ 40 | │ ├── GUIApp.py # GUI APP 41 | │ ├── SortTracker.py # SortImplementation 42 | │ ├── classify_track_count.py # YOLO classifier 43 | | ├── requirements.txt # All required libraries 44 | │ ├── extras 45 | │ |-- coco.data 46 | | |-- coco.names 47 | | |-- multi_classify.cfg 48 | | |-- multi_classify.data 49 | | |-- multi_classify.names 50 | │ |-- yolov4.cfg 51 | | |-- media/Pics_Readme 52 | | |-- GUI.png 53 | | |-- fulldemo.gif 54 | 55 | ## How to run the application 56 | This is the problem with darknet. I can't find a way to give the whole project as a package along with darknet. Please do the following steps extremely carefully. 57 | 1) Clone [AlexyAB's darknet](https://github.com/SravanChittupalli/darknet) github repo. 58 | 2) Run the python demo as given in the [README](https://github.com/AlexeyAB/darknet/blob/master/README.md). If you built and ran the demo successfully then continue to step 3 59 | 3) Clone this repo into the `darknet` folder 60 | 4) Next copy the 3 python files in `Code` folder into the `darknet` folder 61 | 5) Copy the `.data` and `.names` files in `Code/extras` to `darknet/data` folder. 62 | 6) Copy the `.cfg` files in `Code/extras` to `darknet/cfg` folder. 63 | 7) Download the [weight](https://drive.google.com/drive/u/0/folders/1XVWolAhNTvv-ssePnYNXk0GNMrzmwN0w) files from the [drive link](https://drive.google.com/drive/u/0/folders/1XVWolAhNTvv-ssePnYNXk0GNMrzmwN0w) and save them in `darknet` folder. There will also be a `sample_videos` folder also place it anywhere you want. 64 | 8) Open a terminal and run `pip install -r requirements.txt`. I strongly recommend the use of miniconda as is keep the system python packages seperate, thus avoiding conflict. 65 | 9) Now you are all set to run the demo. Activate your environment and run `python GUIApp.py`. 66 | 10) Select the sample video and choose the ROI as shown in the GIF below. Detection of axels requires specific angles so I recommend using the ROI as `(444,191), (1440,734)` to get the best results. 67 | 68 | Just to make your life easier i've added a [bash script]() that you can run to do everything from `step 4` to `step 6` 69 | 70 | ## Demo 71 | ![WORKING DEMO](https://github.com/SravanChittupalli/Advanced-Vehicle-Classifier/blob/master/Code/media/Pics_Readme/fulldemo.gif) 72 | 73 | ## To-Do 74 | I've tried this project using feature extraction , Hough Circle detection , Finding area and length of truck , counting each wheel individually but atlast ended up using 2 iterations of the neural network on 2 different weights one which classifies vehicles and the other that detects wheels :sweat_smile:. I understand that this is not an efficient method but I did not have enough resources to make a whole dataset by myself which includes cars, trucks , bikes , buses along with their wheels. Also I could not find many videos with trck , car facing to the side so that I can detect wheels easily. If anyone has any sugestions on solving this problem efficintly them please open an issue I will be more than happy to try and implement the suggestions or else you can even try to implement it on your own. :smile: 75 | - [x] Add script to ensure smooth running 76 | - [ ] increase # of classes 77 | - [ ] add number plate recognition 78 | - [ ] increase efficiency 79 | 80 | ## References 81 | * [AlexyAB's YOLOV4](https://github.com/AlexeyAB/darknet) 82 | * [SORT Implementation](https://github.com/abewley/sort) 83 | * [Point's side w.r.t a line](https://www.geeksforgeeks.org/direction-point-line-segment/) 84 | * Thanks to pyimagesearch. There is are no topics in CV and ImageProcessing that pyimagesearch blog doesnot cover. 85 | 86 | ## License 87 | Details can be found in [License](LICENSE). 88 | -------------------------------------------------------------------------------- /setup.bash: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env bash 2 | 3 | ###########################Downloading required files############################### 4 | 5 | echo "Copying files..." 6 | 7 | cp Code/classify_track_count.py ../. 8 | cp Code/GUIApp.py ../. 9 | cp Code/SortTracker.py ../. 10 | 11 | cp Code/extras/*.cfg ../cfg 12 | cp Code/extras/*.names ../data 13 | cp Code/extras/*.data ../data 14 | 15 | ######################################################################### --------------------------------------------------------------------------------