├── LICENSE ├── README.md ├── images ├── bus.jpg ├── horses.jpg └── zidane.jpg ├── main.cpp ├── models └── model_here ├── yolo.cpp └── yolo.h /LICENSE: -------------------------------------------------------------------------------- 1 | Apache License 2 | Version 2.0, January 2004 3 | http://www.apache.org/licenses/ 4 | 5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 6 | 7 | 1. Definitions. 8 | 9 | "License" shall mean the terms and conditions for use, reproduction, 10 | and distribution as defined by Sections 1 through 9 of this document. 11 | 12 | "Licensor" shall mean the copyright owner or entity authorized by 13 | the copyright owner that is granting the License. 14 | 15 | "Legal Entity" shall mean the union of the acting entity and all 16 | other entities that control, are controlled by, or are under common 17 | control with that entity. For the purposes of this definition, 18 | "control" means (i) the power, direct or indirect, to cause the 19 | direction or management of such entity, whether by contract or 20 | otherwise, or (ii) ownership of fifty percent (50%) or more of the 21 | outstanding shares, or (iii) beneficial ownership of such entity. 22 | 23 | "You" (or "Your") shall mean an individual or Legal Entity 24 | exercising permissions granted by this License. 25 | 26 | "Source" form shall mean the preferred form for making modifications, 27 | including but not limited to software source code, documentation 28 | source, and configuration files. 29 | 30 | "Object" form shall mean any form resulting from mechanical 31 | transformation or translation of a Source form, including but 32 | not limited to compiled object code, generated documentation, 33 | and conversions to other media types. 34 | 35 | "Work" shall mean the work of authorship, whether in Source or 36 | Object form, made available under the License, as indicated by a 37 | copyright notice that is included in or attached to the work 38 | (an example is provided in the Appendix below). 39 | 40 | "Derivative Works" shall mean any work, whether in Source or Object 41 | form, that is based on (or derived from) the Work and for which the 42 | editorial revisions, annotations, elaborations, or other modifications 43 | represent, as a whole, an original work of authorship. For the purposes 44 | of this License, Derivative Works shall not include works that remain 45 | separable from, or merely link (or bind by name) to the interfaces of, 46 | the Work and Derivative Works thereof. 47 | 48 | "Contribution" shall mean any work of authorship, including 49 | the original version of the Work and any modifications or additions 50 | to that Work or Derivative Works thereof, that is intentionally 51 | submitted to Licensor for inclusion in the Work by the copyright owner 52 | or by an individual or Legal Entity authorized to submit on behalf of 53 | the copyright owner. For the purposes of this definition, "submitted" 54 | means any form of electronic, verbal, or written communication sent 55 | to the Licensor or its representatives, including but not limited to 56 | communication on electronic mailing lists, source code control systems, 57 | and issue tracking systems that are managed by, or on behalf of, the 58 | Licensor for the purpose of discussing and improving the Work, but 59 | excluding communication that is conspicuously marked or otherwise 60 | designated in writing by the copyright owner as "Not a Contribution." 61 | 62 | "Contributor" shall mean Licensor and any individual or Legal Entity 63 | on behalf of whom a Contribution has been received by Licensor and 64 | subsequently incorporated within the Work. 65 | 66 | 2. Grant of Copyright License. Subject to the terms and conditions of 67 | this License, each Contributor hereby grants to You a perpetual, 68 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 69 | copyright license to reproduce, prepare Derivative Works of, 70 | publicly display, publicly perform, sublicense, and distribute the 71 | Work and such Derivative Works in Source or Object form. 72 | 73 | 3. Grant of Patent License. Subject to the terms and conditions of 74 | this License, each Contributor hereby grants to You a perpetual, 75 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 76 | (except as stated in this section) patent license to make, have made, 77 | use, offer to sell, sell, import, and otherwise transfer the Work, 78 | where such license applies only to those patent claims licensable 79 | by such Contributor that are necessarily infringed by their 80 | Contribution(s) alone or by combination of their Contribution(s) 81 | with the Work to which such Contribution(s) was submitted. If You 82 | institute patent litigation against any entity (including a 83 | cross-claim or counterclaim in a lawsuit) alleging that the Work 84 | or a Contribution incorporated within the Work constitutes direct 85 | or contributory patent infringement, then any patent licenses 86 | granted to You under this License for that Work shall terminate 87 | as of the date such litigation is filed. 88 | 89 | 4. Redistribution. You may reproduce and distribute copies of the 90 | Work or Derivative Works thereof in any medium, with or without 91 | modifications, and in Source or Object form, provided that You 92 | meet the following conditions: 93 | 94 | (a) You must give any other recipients of the Work or 95 | Derivative Works a copy of this License; and 96 | 97 | (b) You must cause any modified files to carry prominent notices 98 | stating that You changed the files; and 99 | 100 | (c) You must retain, in the Source form of any Derivative Works 101 | that You distribute, all copyright, patent, trademark, and 102 | attribution notices from the Source form of the Work, 103 | excluding those notices that do not pertain to any part of 104 | the Derivative Works; and 105 | 106 | (d) If the Work includes a "NOTICE" text file as part of its 107 | distribution, then any Derivative Works that You distribute must 108 | include a readable copy of the attribution notices contained 109 | within such NOTICE file, excluding those notices that do not 110 | pertain to any part of the Derivative Works, in at least one 111 | of the following places: within a NOTICE text file distributed 112 | as part of the Derivative Works; within the Source form or 113 | documentation, if provided along with the Derivative Works; or, 114 | within a display generated by the Derivative Works, if and 115 | wherever such third-party notices normally appear. The contents 116 | of the NOTICE file are for informational purposes only and 117 | do not modify the License. You may add Your own attribution 118 | notices within Derivative Works that You distribute, alongside 119 | or as an addendum to the NOTICE text from the Work, provided 120 | that such additional attribution notices cannot be construed 121 | as modifying the License. 122 | 123 | You may add Your own copyright statement to Your modifications and 124 | may provide additional or different license terms and conditions 125 | for use, reproduction, or distribution of Your modifications, or 126 | for any such Derivative Works as a whole, provided Your use, 127 | reproduction, and distribution of the Work otherwise complies with 128 | the conditions stated in this License. 129 | 130 | 5. Submission of Contributions. Unless You explicitly state otherwise, 131 | any Contribution intentionally submitted for inclusion in the Work 132 | by You to the Licensor shall be under the terms and conditions of 133 | this License, without any additional terms or conditions. 134 | Notwithstanding the above, nothing herein shall supersede or modify 135 | the terms of any separate license agreement you may have executed 136 | with Licensor regarding such Contributions. 137 | 138 | 6. Trademarks. This License does not grant permission to use the trade 139 | names, trademarks, service marks, or product names of the Licensor, 140 | except as required for reasonable and customary use in describing the 141 | origin of the Work and reproducing the content of the NOTICE file. 142 | 143 | 7. Disclaimer of Warranty. Unless required by applicable law or 144 | agreed to in writing, Licensor provides the Work (and each 145 | Contributor provides its Contributions) on an "AS IS" BASIS, 146 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or 147 | implied, including, without limitation, any warranties or conditions 148 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A 149 | PARTICULAR PURPOSE. You are solely responsible for determining the 150 | appropriateness of using or redistributing the Work and assume any 151 | risks associated with Your exercise of permissions under this License. 152 | 153 | 8. Limitation of Liability. In no event and under no legal theory, 154 | whether in tort (including negligence), contract, or otherwise, 155 | unless required by applicable law (such as deliberate and grossly 156 | negligent acts) or agreed to in writing, shall any Contributor be 157 | liable to You for damages, including any direct, indirect, special, 158 | incidental, or consequential damages of any character arising as a 159 | result of this License or out of the use or inability to use the 160 | Work (including but not limited to damages for loss of goodwill, 161 | work stoppage, computer failure or malfunction, or any and all 162 | other commercial damages or losses), even if such Contributor 163 | has been advised of the possibility of such damages. 164 | 165 | 9. Accepting Warranty or Additional Liability. While redistributing 166 | the Work or Derivative Works thereof, You may choose to offer, 167 | and charge a fee for, acceptance of support, warranty, indemnity, 168 | or other liability obligations and/or rights consistent with this 169 | License. However, in accepting such obligations, You may act only 170 | on Your own behalf and on Your sole responsibility, not on behalf 171 | of any other Contributor, and only if You agree to indemnify, 172 | defend, and hold each Contributor harmless for any liability 173 | incurred by, or claims asserted against, such Contributor by reason 174 | of your accepting any such warranty or additional liability. 175 | 176 | END OF TERMS AND CONDITIONS 177 | 178 | APPENDIX: How to apply the Apache License to your work. 179 | 180 | To apply the Apache License to your work, attach the following 181 | boilerplate notice, with the fields enclosed by brackets "[]" 182 | replaced with your own identifying information. (Don't include 183 | the brackets!) The text should be enclosed in the appropriate 184 | comment syntax for the file format. We also recommend that a 185 | file or class name and description of purpose be included on the 186 | same "printed page" as the copyright notice for easier 187 | identification within third-party archives. 188 | 189 | Copyright [yyyy] [name of copyright owner] 190 | 191 | Licensed under the Apache License, Version 2.0 (the "License"); 192 | you may not use this file except in compliance with the License. 193 | You may obtain a copy of the License at 194 | 195 | http://www.apache.org/licenses/LICENSE-2.0 196 | 197 | Unless required by applicable law or agreed to in writing, software 198 | distributed under the License is distributed on an "AS IS" BASIS, 199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 200 | See the License for the specific language governing permissions and 201 | limitations under the License. 202 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # yolov7-opencv-dnn-cpp 2 | 使用opencv模块部署yolov7-0.1版本和yolov5-6.0以上版本
3 | 4 | + 基于yolov5-6.0版本的yolov5:https://github.com/ultralytics/yolov5
5 | + 基于yolov7-0.1的版本https://github.com/WongKinYiu/yolov7
6 | 7 | ## **OpenCV>=4.5.0**
8 | 9 | > python path/to/export.py --weights yolov5s.pt --img [640,640] --opset 12 --include onnx
10 | > python path/to/export.py --weights yolov7.pt --img [640,640]
11 | 请注意，yolov7导出的时候不要加--grid这个参数(控制detect层),否则opencv读取没问题，推理报错.
12 | 13 | 可以通过yolo.h中定义的YOLOV5宏定义来切换yolov5和yolov7两个版本，(其实两个版本onnx后处理方式差不多的说
14 | >通过yolo.h中定义的YOLO_P6来切换是否使用两者的P6模型。
15 | > YOLOV5:true -->yolov5.onnx
16 | > YOLOV5:false-->yolov7.onnx 17 | #### 2022.12.13 更新： 18 | + 如果你的显卡支持FP16推理的话，可以将模型读取代码中的```DNN_TARGET_CUDA```改成```DNN_TARGET_CUDA_FP16```提升推理速度（虽然是蚊子腿，好歹也是肉（： 19 | #### 2022-09-06 更新: 20 | + 最近有些小伙伴使用opencv4.6的版本报错了，经过debug发现，opencv4.6的和4.5.x的forward输出顺序不一样导致的，使用opencv4.6的时候在net.forward之后需要加上一个排序，使得输出口从大到小排序才行。
21 | https://github.com/UNeedCryDear/yolov7-opencv-dnn-cpp/blob/86d4f5ef6ecfd7eb36a14d0c06a84a5468ff98e6/yolo.cpp#L48 22 | #### 2022-10-18 更新: 23 | yolov7目前有些模型低于opencv4.5.5会报错,报错信息类似下面使用opencv4.5.0读取yolov7-d6.pt转出的onnx模型（不能加参数--grid），此时建议升级下opencv的版本 24 | > >OpenCV(4.5.0) Error: Unspecified error (> Node [Slice]:(341) parse error: OpenCV(4.5.0) D:\opencv\ocv4.5.0\sources\modules\dnn\src\onnx\onnx_importer.cpp:697: error: (-2:Unspecified error) in function 'void __cdecl cv::dnn::dnn4_v20200908::ONNXImporter::handleNode(const class opencv_onnx::NodeProto &)'
25 | > > Slice layer only supports steps = 1 (expected: 'countNonZero(step_blob != 1) == 0'), where
26 | > > 'countNonZero(step_blob != 1)' is 1
27 | > > must be equal to
28 | > > '0' is 0
29 | > > in cv::dnn::dnn4_v20200908::ONNXImporter::handleNode, file D:\opencv\ocv4.5.0\sources\modules\dnn\src\onnx\onnx_importer.cpp, line 1797
30 | 31 | debug可以发现是由于yolov7-d6中使用了ReOrg模块引起的报错，这个模块有点类似早期的yolov5的Facos模块，如果一定要在opencv4.5.0下面运行，需要将ReOrg模块修改成下面的代码。 32 | 在models/common.py里面搜索下ReOrg. 33 | ``` 34 | class ReOrg(nn.Module): 35 | def __init__(self): 36 | super(ReOrg, self).__init__() 37 | 38 | def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2) 39 | #origin code 40 | # return torch.cat([x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]], 1) 41 | self.concat=Contract(gain=2) 42 | return self.concat(x) 43 | ``` 44 | 45 | 46 | 47 | 另外关于换行符，windows下面需要设置为CRLF，上传到github会自动切换成LF，windows下面切换一下即可。
48 | 贴个yolov7.onnx和yolov5s.onnx的对比
49 | ![yolo](https://user-images.githubusercontent.com/52729998/180824922-0c7dc3f9-fbda-497b-9ae3-3f299b8c0452.png) 50 | -------------------------------------------------------------------------------- /images/bus.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/UNeedCryDear/yolov7-opencv-dnn-cpp/7547f390164d3f558893ecfb68cb608918e72020/images/bus.jpg -------------------------------------------------------------------------------- /images/horses.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/UNeedCryDear/yolov7-opencv-dnn-cpp/7547f390164d3f558893ecfb68cb608918e72020/images/horses.jpg -------------------------------------------------------------------------------- /images/zidane.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/UNeedCryDear/yolov7-opencv-dnn-cpp/7547f390164d3f558893ecfb68cb608918e72020/images/zidane.jpg -------------------------------------------------------------------------------- /main.cpp: -------------------------------------------------------------------------------- 1 | #include "yolo.h" 2 | #include 3 | #include 4 | #include 5 | 6 | #define USE_CUDA true //use opencv-cuda 7 | 8 | using namespace std; 9 | using namespace cv; 10 | using namespace dnn; 11 | 12 | 13 | int main() 14 | { 15 | string img_path = "./images/bus.jpg"; 16 | 17 | #if(defined YOLOV5 && YOLOV5==true) 18 | string model_path = "models/yolov5s.onnx"; 19 | #else 20 | string model_path = "models/yolov7.onnx"; 21 | #endif 22 | 23 | 24 | Yolo test; 25 | Net net; 26 | if (test.readModel(net, model_path, USE_CUDA)) { 27 | cout << "read net ok!" << endl; 28 | } 29 | else { 30 | cout << "read onnx model failed!"; 31 | return -1; 32 | } 33 | 34 | //生成随机颜色 35 | vector color; 36 | srand(time(0)); 37 | for (int i = 0; i < 80; i++) { 38 | int b = rand() % 256; 39 | int g = rand() % 256; 40 | int r = rand() % 256; 41 | color.push_back(Scalar(b, g, r)); 42 | } 43 | vector result; 44 | Mat img = imread(img_path); 45 | 46 | if (test.Detect(img, net, result)) { 47 | test.drawPred(img, result, color); 48 | 49 | } 50 | else { 51 | cout << "Detect Failed!" << endl; 52 | } 53 | 54 | system("pause"); 55 | return 0; 56 | } 57 | -------------------------------------------------------------------------------- /models/model_here: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /yolo.cpp: -------------------------------------------------------------------------------- 1 | #include"yolo.h" 2 | using namespace std; 3 | using namespace cv; 4 | using namespace cv::dnn; 5 | 6 | bool Yolo::readModel(Net& net, string& netPath, bool isCuda = false) { 7 | try { 8 | net = readNet(netPath); 9 | } 10 | catch (const std::exception&) { 11 | return false; 12 | } 13 | //cuda 14 | if (isCuda) { 15 | net.setPreferableBackend(cv::dnn::DNN_BACKEND_CUDA); 16 | net.setPreferableTarget(cv::dnn::DNN_TARGET_CUDA); 17 | } 18 | //cpu 19 | else { 20 | net.setPreferableBackend(cv::dnn::DNN_BACKEND_DEFAULT); 21 | net.setPreferableTarget(cv::dnn::DNN_TARGET_CPU); 22 | } 23 | return true; 24 | } 25 | #if(defined YOLOV5 && YOLOV5==false) //yolov7 26 | bool Yolo::Detect(Mat& SrcImg, Net& net, vector& output) { 27 | Mat blob; 28 | int col = SrcImg.cols; 29 | int row = SrcImg.rows; 30 | int maxLen = MAX(col, row); 31 | Mat netInputImg = SrcImg.clone(); 32 | if (maxLen > 1.2 * col || maxLen > 1.2 * row) { 33 | Mat resizeImg = Mat::zeros(maxLen, maxLen, CV_8UC3); 34 | SrcImg.copyTo(resizeImg(Rect(0, 0, col, row))); 35 | netInputImg = resizeImg; 36 | } 37 | vector > layer; 38 | vector layer_names; 39 | layer_names= net.getLayerNames(); 40 | blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(0, 0, 0), true, false); 41 | //如果在其他设置没有问题的情况下但是结果偏差很大，可以尝试下用下面两句语句 42 | //blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(104, 117, 123), true, false); 43 | //blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(114, 114,114), true, false); 44 | net.setInput(blob); 45 | std::vector netOutputImg; 46 | net.forward(netOutputImg, net.getUnconnectedOutLayersNames()); 47 | #if CV_VERSION_MAJOR==4&&CV_VERSION_MINOR==6 48 | std::sort(netOutputImg.begin(), netOutputImg.end(), [](Mat &A, Mat &B) {return A.size[2] > B.size[2]; });//opencv 4.6 49 | #endif 50 | std::vector classIds;//结果id数组 51 | std::vector confidences;//结果每个id对应置信度数组 52 | std::vector boxes;//每个id矩形框 53 | float ratio_h = (float)netInputImg.rows / netHeight; 54 | float ratio_w = (float)netInputImg.cols / netWidth; 55 | int net_width = className.size() + 5; //输出的网络宽度是类别数+5 56 | for (int stride = 0; stride < strideSize; stride++) { //stride 57 | float* pdata = (float*)netOutputImg[stride].data; 58 | int grid_x = (int)(netWidth / netStride[stride]); 59 | int grid_y = (int)(netHeight / netStride[stride]); 60 | for (int anchor = 0; anchor < 3; anchor++) { //anchors 61 | const float anchor_w = netAnchors[stride][anchor * 2]; 62 | const float anchor_h = netAnchors[stride][anchor * 2 + 1]; 63 | for (int i = 0; i < grid_y; i++) { 64 | for (int j = 0; j < grid_x; j++) { 65 | float box_score = sigmoid_x(pdata[4]); ;//获取每一行的box框中含有某个物体的概率 66 | if (box_score >= boxThreshold) { 67 | cv::Mat scores(1, className.size(), CV_32FC1, pdata + 5); 68 | Point classIdPoint; 69 | double max_class_socre; 70 | minMaxLoc(scores, 0, &max_class_socre, 0, &classIdPoint); 71 | max_class_socre = sigmoid_x(max_class_socre); 72 | if (max_class_socre >= classThreshold) { 73 | float x = (sigmoid_x(pdata[0]) * 2.f - 0.5f + j) * netStride[stride]; //x 74 | float y = (sigmoid_x(pdata[1]) * 2.f - 0.5f + i) * netStride[stride]; //y 75 | float w = powf(sigmoid_x(pdata[2]) * 2.f, 2.f) * anchor_w; //w 76 | float h = powf(sigmoid_x(pdata[3]) * 2.f, 2.f) * anchor_h; //h 77 | int left = (int)(x - 0.5 * w) * ratio_w + 0.5; 78 | int top = (int)(y - 0.5 * h) * ratio_h + 0.5; 79 | classIds.push_back(classIdPoint.x); 80 | confidences.push_back(max_class_socre * box_score); 81 | boxes.push_back(Rect(left, top, int(w * ratio_w), int(h * ratio_h))); 82 | } 83 | } 84 | pdata += net_width;//下一行 85 | } 86 | } 87 | } 88 | } 89 | 90 | //执行非最大抑制以消除具有较低置信度的冗余重叠框（NMS） 91 | vector nms_result; 92 | NMSBoxes(boxes, confidences, nmsScoreThreshold, nmsThreshold, nms_result); 93 | for (int i = 0; i < nms_result.size(); i++) { 94 | int idx = nms_result[i]; 95 | Output result; 96 | result.id = classIds[idx]; 97 | result.confidence = confidences[idx]; 98 | result.box = boxes[idx]; 99 | output.push_back(result); 100 | } 101 | if (output.size()) 102 | return true; 103 | else 104 | return false; 105 | } 106 | #else 107 | //yolov5 108 | bool Yolo::Detect(Mat& SrcImg, Net& net, vector& output) { 109 | Mat blob; 110 | int col = SrcImg.cols; 111 | int row = SrcImg.rows; 112 | int maxLen = MAX(col, row); 113 | Mat netInputImg = SrcImg.clone(); 114 | if (maxLen > 1.2 * col || maxLen > 1.2 * row) { 115 | Mat resizeImg = Mat::zeros(maxLen, maxLen, CV_8UC3); 116 | SrcImg.copyTo(resizeImg(Rect(0, 0, col, row))); 117 | netInputImg = resizeImg; 118 | } 119 | vector > layer; 120 | vector layer_names; 121 | layer_names = net.getLayerNames(); 122 | blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(0, 0, 0), true, false); 123 | //如果在其他设置没有问题的情况下但是结果偏差很大，可以尝试下用下面两句语句 124 | //blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(104, 117, 123), true, false); 125 | //blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(114, 114,114), true, false); 126 | net.setInput(blob); 127 | std::vector netOutputImg; 128 | net.forward(netOutputImg, net.getUnconnectedOutLayersNames()); 129 | std::vector classIds;//结果id数组 130 | std::vector confidences;//结果每个id对应置信度数组 131 | std::vector boxes;//每个id矩形框 132 | float ratio_h = (float)netInputImg.rows / netHeight; 133 | float ratio_w = (float)netInputImg.cols / netWidth; 134 | int net_width = className.size() + 5; //输出的网络宽度是类别数+5 135 | float* pdata = (float*)netOutputImg[0].data; 136 | for (int stride = 0; stride < strideSize; stride++) { //stride 137 | 138 | int grid_x = (int)(netWidth / netStride[stride]); 139 | int grid_y = (int)(netHeight / netStride[stride]); 140 | for (int anchor = 0; anchor < 3; anchor++) { //anchors 141 | //const float anchor_w = netAnchors[stride][anchor * 2]; 142 | //const float anchor_h = netAnchors[stride][anchor * 2 + 1]; 143 | for (int i = 0; i < grid_y; i++) { 144 | for (int j = 0; j < grid_x; j++) { 145 | float box_score =pdata[4]; ;//获取每一行的box框中含有某个物体的概率 146 | if (box_score >= boxThreshold) { 147 | cv::Mat scores(1, className.size(), CV_32FC1, pdata + 5); 148 | Point classIdPoint; 149 | double max_class_socre; 150 | minMaxLoc(scores, 0, &max_class_socre, 0, &classIdPoint); 151 | max_class_socre =max_class_socre; 152 | if (max_class_socre >= classThreshold) { 153 | float x = pdata[0];// (sigmoid_x(pdata[0]) * 2.f - 0.5f + j) * m_Mark_Stride[stride]; //x 154 | float y = pdata[1];//(sigmoid_x(pdata[1]) * 2.f - 0.5f + i) * m_Mark_Stride[stride]; //y 155 | float w = pdata[2];//powf(sigmoid_x(pdata[2]) * 2.f, 2.f) * anchor_w*ratio_c; //w 156 | float h = pdata[3];//powf(sigmoid_x(pdata[3]) * 2.f, 2.f) * anchor_h*ratio_r; //h 157 | int left = round((x - 0.5 * w) * ratio_w) ; 158 | int top = round((y - 0.5 * h) * ratio_h); 159 | classIds.push_back(classIdPoint.x); 160 | confidences.push_back(max_class_socre * box_score); 161 | boxes.push_back(Rect(left, top, round(w * ratio_w), round(h * ratio_h))); 162 | } 163 | } 164 | pdata += net_width;//下一行 165 | } 166 | } 167 | } 168 | } 169 | 170 | //执行非最大抑制以消除具有较低置信度的冗余重叠框（NMS） 171 | vector nms_result; 172 | NMSBoxes(boxes, confidences, nmsScoreThreshold, nmsThreshold, nms_result); 173 | for (int i = 0; i < nms_result.size(); i++) { 174 | int idx = nms_result[i]; 175 | Output result; 176 | result.id = classIds[idx]; 177 | result.confidence = confidences[idx]; 178 | result.box = boxes[idx]; 179 | output.push_back(result); 180 | } 181 | if (output.size()) 182 | return true; 183 | else 184 | return false; 185 | } 186 | #endif 187 | void Yolo::drawPred(Mat& img, vector result, vector color) { 188 | for (int i = 0; i < result.size(); i++) { 189 | int left, top; 190 | left = result[i].box.x; 191 | top = result[i].box.y; 192 | int color_num = i; 193 | rectangle(img, result[i].box, color[result[i].id], 2, 8); 194 | 195 | string label = className[result[i].id] + ":" + to_string(result[i].confidence); 196 | 197 | int baseLine; 198 | Size labelSize = getTextSize(label, FONT_HERSHEY_SIMPLEX, 0.5, 1, &baseLine); 199 | top = max(top, labelSize.height); 200 | //rectangle(frame, Point(left, top - int(1.5 * labelSize.height)), Point(left + int(1.5 * labelSize.width), top + baseLine), Scalar(0, 255, 0), FILLED); 201 | putText(img, label, Point(left, top), FONT_HERSHEY_SIMPLEX, 1, color[result[i].id], 2); 202 | } 203 | imshow("1", img); 204 | //imwrite("out.bmp", img); 205 | waitKey(); 206 | //destroyAllWindows(); 207 | 208 | } 209 | -------------------------------------------------------------------------------- /yolo.h: -------------------------------------------------------------------------------- 1 | #pragma once 2 | #include 3 | #include 4 | 5 | #ifndef YOLOV5 6 | #define YOLOV5 false //true:Yolov5, false:yolov7 7 | #endif 8 | 9 | #ifndef YOLO_P6 10 | #define YOLO_P6 false //是否使用P6模型 11 | #endif 12 | 13 | 14 | 15 | struct Output { 16 | int id; //结果类别id 17 | float confidence; //结果置信度 18 | cv::Rect box; //矩形框 19 | }; 20 | 21 | class Yolo { 22 | public: 23 | Yolo() { 24 | } 25 | ~Yolo() {} 26 | bool readModel(cv::dnn::Net& net, std::string& netPath, bool isCuda); 27 | bool Detect(cv::Mat& SrcImg, cv::dnn::Net& net, std::vector& output); 28 | void drawPred(cv::Mat& img, std::vector result, std::vector color); 29 | 30 | private: 31 | 32 | float sigmoid_x(float x) 33 | { 34 | return static_cast(1.f / (1.f + exp(-x))); 35 | } 36 | #if(defined YOLO_P6 && YOLO_P6==true) 37 | 38 | #if(defined YOLOV5 && YOLOV5==false) 39 | const float netAnchors[4][6] = { { 19,27, 44,40, 38,94 },{96,68, 86,152, 180,137} ,{140,301, 303,264, 238,542}, { 436,615, 739,380, 925,792 } };//yolov7-P6 anchors 40 | #else 41 | const float netAnchors[4][6]= { { 19,27, 44,40, 38,94 },{ 96,68, 86,152, 180,137 },{ 140,301, 303,264, 238,542 },{ 436,615, 739,380, 925,792 } }; //yolov5-P6 anchors 42 | #endif 43 | const int netWidth = 1280; //ONNX图片输入宽度 44 | const int netHeight = 1280; //ONNX图片输入高度 45 | const int strideSize = 4; //stride size 46 | #else 47 | #if(defined YOLOV5 && YOLOV5==false) 48 | const float netAnchors[3][6] = { {12, 16, 19, 36, 40, 28},{36, 75, 76, 55, 72, 146},{142, 110, 192, 243, 459, 401} }; //yolov7-P5 anchors 49 | #else 50 | const float netAnchors[3][6] = { { 10, 13, 16, 30, 33, 23 }, { 30, 61, 62, 45, 59, 119 }, { 116, 90, 156, 198, 373, 326 } };//yolov5-P5 anchors 51 | #endif 52 | const int netWidth = 640; //ONNX图片输入宽度 53 | const int netHeight = 640; //ONNX图片输入高度 54 | const int strideSize = 3; //stride size 55 | #endif // YOLO_P6 56 | 57 | const float netStride[4] = { 8, 16.0,32,64 }; 58 | 59 | float boxThreshold = 0.25; 60 | float classThreshold = 0.25; 61 | float nmsThreshold = 0.45; 62 | float nmsScoreThreshold = boxThreshold * classThreshold; 63 | 64 | std::vector className = { "person", "bicycle", "car", "motorcycle", "airplane", "bus", "train", "truck", "boat", "traffic light", 65 | "fire hydrant", "stop sign", "parking meter", "bench", "bird", "cat", "dog", "horse", "sheep", "cow", 66 | "elephant", "bear", "zebra", "giraffe", "backpack", "umbrella", "handbag", "tie", "suitcase", "frisbee", 67 | "skis", "snowboard", "sports ball", "kite", "baseball bat", "baseball glove", "skateboard", "surfboard", 68 | "tennis racket", "bottle", "wine glass", "cup", "fork", "knife", "spoon", "bowl", "banana", "apple", 69 | "sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza", "donut", "cake", "chair", "couch", 70 | "potted plant", "bed", "dining table", "toilet", "tv", "laptop", "mouse", "remote", "keyboard", "cell phone", 71 | "microwave", "oven", "toaster", "sink", "refrigerator", "book", "clock", "vase", "scissors", "teddy bear", 72 | "hair drier", "toothbrush" }; 73 | }; 74 | --------------------------------------------------------------------------------