一、环境配置
1.1 需要编译gpu版的opencv：
（1）下载cmake，opencv，opencv_contrib，Visual Studio
（2）使用cmake编译opencv
注意：编译时可以会出现下载不了文件，可通过git加速的方式进行下载，并放置到相应目录
（3）编译好之后，在vs中进行生成。
1.2 需要安装yolov5环境，并训练自己的模型
（1）anconda环境搭建
conda create -n yolo python=3.6
亲测，使用3.6版本比较好，其他版本各种报错
（2）yolov5下载
github官网下载https://github.com/ultralytics/yolov5
（3）素材整理
整理自己的数据集
（4）模型训练
python train.py --img 640 --batch 50 --epochs 100 --data …/yolo_A/A.yaml --weights yolov5s.pt --nosave --cache
（5）效果预测
python detect.py --weights /content/yolov5/runs/train/exp2/weights/best.pt --img 640 --conf 0.25 --source …/test.jpg
二、C++部署代码
main.cpp
#include "stdafx.h"
#include "yolo.h"
#include <iostream>
#include<opencv2//opencv.hpp>
#include<math.h>
using namespace std;
using namespace cv;
using namespace dnn;
int main()
{
string img_path = "./image/bus.jpg";
string model_path = "yolov5s.onnx";
//int num_devices = cv::cuda::getCudaEnabledDeviceCount();
//if (num_devices <= 0) {
//cerr << "There is no cuda." << endl;
//return -1;
//}
//else {
//cout << num_devices << endl;
//}
Yolo test;
Net net;
if (test.readModel(net, model_path, false)) {
cout << "read net ok!" << endl;
}
else {
return -1;
}
//生成随机颜色
vector<Scalar> color;
srand(time(0));
for (int i = 0; i < 80; i++) {
int b = rand() % 256;
int g = rand() % 256;
int r = rand() % 256;
color.push_back(Scalar(b, g, r));
}
vector<Output> result;
Mat img = imread(img_path);
if (test.Detect(img, net, result)) {
test.drawPred(img, result, color);
}
else {
cout << "Detect Failed!"<<endl;
}
system("pause");
return 0;
}
yolo.cpp
#include"stdafx.h"
#include"yolo.h";
using namespace std;
using namespace cv;
using namespace dnn;
bool Yolo::readModel(Net &net, string &netPath,bool isCuda = false) {
try {
net = readNet(netPath);
}
catch (const std::exception&) {
return false;
}
//cuda
if (isCuda) {
net.setPreferableBackend(cv::dnn::DNN_BACKEND_CUDA);
net.setPreferableTarget(cv::dnn::DNN_TARGET_CUDA_FP16);
}
//cpu
else {
net.setPreferableBackend(cv::dnn::DNN_BACKEND_DEFAULT);
net.setPreferableTarget(cv::dnn::DNN_TARGET_CPU);
}
return true;
}
bool Yolo::Detect(Mat &SrcImg, Net &net, vector<Output> &output) {
Mat blob;
int col = SrcImg.cols;
int row = SrcImg.rows;
int maxLen = MAX(col, row);
Mat netInputImg = SrcImg.clone();
if (maxLen > 1.2*col || maxLen > 1.2*row) {
Mat resizeImg = Mat::zeros(maxLen, maxLen, CV_8UC3);
SrcImg.copyTo(resizeImg(Rect(0, 0, col, row)));
netInputImg = resizeImg;
}
blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(104, 117,123), true, false);
//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
//blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(0, 0,0), true, false);
//blobFromImage(netInputImg, blob, 1 / 255.0, cv::Size(netWidth, netHeight), cv::Scalar(114, 114,114), true, false);
net.setInput(blob);
std::vector<cv::Mat> netOutputImg;
//vector<string> outputLayerName{"345","403", "461","output" };
//net.forward(netOutputImg, outputLayerName[3]); //获取output的输出
net.forward(netOutputImg, net.getUnconnectedOutLayersNames());
std::vector<int> classIds;//结果id数组
std::vector<float> confidences;//结果每个id对应置信度数组
std::vector<cv::Rect> boxes;//每个id矩形框
float ratio_h = (float)netInputImg.rows / netHeight;
float ratio_w = (float)netInputImg.cols / netWidth;
int net_width = className.size() + 5; //输出的网络宽度是类别数+5
float* pdata = (float*)netOutputImg[0].data;
for (int stride =0; stride < 3; stride++) { //stride
int grid_x = (int)(netWidth / netStride[stride]);
int grid_y = (int)(netHeight / netStride[stride]);
for (int anchor = 0; anchor < 3; anchor++) { //anchors
const float anchor_w = netAnchors[stride][anchor * 2];
const float anchor_h = netAnchors[stride][anchor * 2 + 1];
for (int i = 0; i < grid_y; i++) {
for (int j = 0; j < grid_x; j++) {
float box_score = pdata[4]; //Sigmoid(pdata[4]);//获取每一行的box框中含有某个物体的概率
if (box_score > boxThreshold) {
cv::Mat scores(1, className.size(), CV_32FC1, pdata + 5);
Point classIdPoint;
double max_class_socre;
minMaxLoc(scores, 0, &max_class_socre, 0, &classIdPoint);
max_class_socre = (float)max_class_socre; //Sigmoid((float)max_class_socre);
if (max_class_socre > classThreshold) {
//rect [x,y,w,h]
float x = pdata[0];// (Sigmoid(pdata[0]) * 2.f - 0.5f + j) * netStride[stride]; //x
float y = pdata[1];// (Sigmoid(pdata[1]) * 2.f - 0.5f + i) * netStride[stride]; //y
float w = pdata[2];// powf(Sigmoid(pdata[2]) * 2.f, 2.f) * anchor_w; //w
float h = pdata[3];// powf(Sigmoid(pdata[3]) * 2.f, 2.f) * anchor_h; //h
int left = (x - 0.5*w)*ratio_w;
int top = (y - 0.5*h)*ratio_h;
classIds.push_back(classIdPoint.x);
confidences.push_back(max_class_socre*box_score);
boxes.push_back(Rect(left, top, int(w*ratio_w), int(h*ratio_h)));
}
}
pdata += net_width;//下一行
}
}
}
}
//执行非最大抑制以消除具有较低置信度的冗余重叠框（NMS）
vector<int> nms_result;
NMSBoxes(boxes, confidences, classThreshold, nmsThreshold, nms_result);
for (int i = 0; i < nms_result.size(); i++) {
int idx = nms_result[i];
Output result;
result.id = classIds[idx];
result.confidence = confidences[idx];
result.box = boxes[idx];
output.push_back(result);
}
if (output.size())
return true;
else
return false;
}
void Yolo::drawPred(Mat &img, vector<Output> result, vector<Scalar> color) {
for (int i = 0; i < result.size(); i++) {
int left, top;
left = result[i].box.x;
top = result[i].box.y;
int color_num = i;
rectangle(img, result[i].box, color[result[i].id], 2, 8);
string label = className[result[i].id] +":" + to_string(result[i].confidence);
int baseLine;
Size labelSize = getTextSize(label, FONT_HERSHEY_SIMPLEX, 0.5, 1, &baseLine);
top = max(top, labelSize.height);
//rectangle(frame, Point(left, top - int(1.5 * labelSize.height)), Point(left + int(1.5 * labelSize.width), top + baseLine), Scalar(0, 255, 0), FILLED);
putText(img, label, Point(left, top), FONT_HERSHEY_SIMPLEX, 1, color[result[i].id], 2);
}
imshow("1", img);
//imwrite("out.bmp", img);
waitKey();
//destroyAllWindows();
}
yolo.h
#pragma once
#include<iostream>
#include<opencv2/opencv.hpp>
//#include<string>
//using namespace std;
//using namespace cv;
//using namespace dnn;
struct Output {
int id;//结果类别id
float confidence;//结果置信度
cv::Rect box;//矩形框
};
class Yolo {
public:
Yolo() {
}
~Yolo() {}
bool readModel(cv::dnn::Net &net,std::string &netPath,bool isCuda);
bool Detect(cv::Mat &SrcImg,cv::dnn::Net &net, std::vector<Output> &output);
void drawPred(cv::Mat &img, std::vector<Output> result, std::vector<cv::Scalar> color);
private:
const float netAnchors[3][6] = { { 10.0, 13.0, 16.0, 30.0, 33.0, 23.0 },{ 30.0, 61.0, 62.0, 45.0, 59.0, 119.0 },{ 116.0, 90.0, 156.0, 198.0, 373.0, 326.0 } };
const float netStride[3] = { 8, 16.0,32 };
const int netWidth = 640;
const int netHeight = 640;
float nmsThreshold = 0.45;
float boxThreshold = 0.25;
float classThreshold = 0.25;
std::vector<std::string> className = { "person", "bicycle", "car", "motorcycle", "airplane", "bus", "train", "truck", "boat", "traffic light",
"fire hydrant", "stop sign", "parking meter", "bench", "bird", "cat", "dog", "horse", "sheep", "cow",
"elephant", "bear", "zebra", "giraffe", "backpack", "umbrella", "handbag", "tie", "suitcase", "frisbee",
"skis", "snowboard", "sports ball", "kite", "baseball bat", "baseball glove", "skateboard", "surfboard",
"tennis racket", "bottle", "wine glass", "cup", "fork", "knife", "spoon", "bowl", "banana", "apple",
"sandwich", "orange", "broccoli", "carrot", "hot dog", "pizza", "donut", "cake", "chair", "couch",
"potted plant", "bed", "dining table", "toilet", "tv", "laptop", "mouse", "remote", "keyboard", "cell phone",
"microwave", "oven", "toaster", "sink", "refrigerator", "book", "clock", "vase", "scissors", "teddy bear",
"hair drier", "toothbrush" };
};
4、gpu加速语句
如果自己写程序，记得加入这条语句：
net.setPreferableBackend(cv::dnn::DNN_BACKEND_CUDA);
net.setPreferableTarget(cv::dnn::DNN_TARGET_CUDA);
参考文献：https://blog.csdn.net/playezio/article/details/123652814