OpenCV DNN模块官方教程地址如下，可以查看各个对应的使用方法https://docs.opencv.org/4.4.0/d2/d58/tutorial_table_of_content_dnn.html
今天介绍第五部分：加载darknet框架的YoloV4模型做目标检测，相较于官方文档更易理解，之所以选YoloV4，是因为YoloV4现已很流行，同时YoloV4和YoloV3在OpenCV DNN模块的使用方法相似，下面的代码只需要改动YoloV3对应的权重和配置文件就可以。
代码语言：javascript复制String config = "./model/yolov4.cfg";
String weights = "./model/yolov4.weights";
string classesFile = "./model/coco.names";上面三个文件的下载地址分别是：
https://raw.githubusercontent.com/AlexeyAB/darknet/master/cfg/yolov4.cfg
https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v3_optimal/yolov4.weights
https://github.com/pjreddie/darknet/blob/master/data/coco.names
代码语言：javascript复制// 加载darknet网络
Net net = readNetFromDarknet(config, weights);OpenCV DNN模块支持常见深度学习框架如TensorFlowCaffe、Darknet等，对应的函数：readNetFromTensorflow、readNetFromCaffe.
下面是OpenCV DNN读取YoloV4模型进行图片检测代码和效果：
代码语言：javascript复制// DNN_YOLO_V4.cpp : 此文件包含 "main" 函数。程序执行将在此处开始并结束。
#include "pch.h"
#include<opencv2/opencv.hpp>
#include<opencv2/dnn.hpp>
#include <iostream>
#include <fstream>
#include<istream>
#include<string>
using namespace std;
using namespace cv;
using namespace dnn;
// 初始化参数
float confThreshold = 0.5; // 置信度阈值
float nmsThreshold = 0.4; // 非极大值抑制(NMS)阈值
int inpWidth = 416; // 网络输入图像宽度
int inpHeight = 416; // 网络输入图像高度
// 加载类别名称文件
vector<string>classes;
// Load names of classes
string classesFile = "./model/coco.names";
ifstream ifs(classesFile.c_str());
// 设置模型配置文件和权重
String config = "./model/yolov4.cfg";
String weights = "./model/yolov4.weights";
// 加载网络
Net net = readNetFromDarknet(config, weights);
// 获取输出层名称
vector<String> getOutputsNames(const Net& net)
{
static vector<String> names;
if (names.empty())
{
//Get the indices of the output layers, i.e. the layers with unconnected outputs
vector<int> outLayers = net.getUnconnectedOutLayers();
//get the names of all the layers in the network
vector<String> layersNames = net.getLayerNames();
// Get the names of the output layers in names
names.resize(outLayers.size());
for (size_t i = 0; i < outLayers.size(); i)
names[i] = layersNames[outLayers[i] - 1];
}
return names;
}
// Draw the predicted bounding box
void drawPred(int classId, float conf, int left, int top, int right, int bottom, Mat& frame)
{
//Draw a rectangle displaying the bounding box
rectangle(frame, Point(left, top), Point(right, bottom), Scalar(0, 0, 255), 2);
//Get the label for the class name and its confidence
string label = format("%.2f", conf);
if (!classes.empty())
{
CV_Assert(classId < (int)classes.size());
label = classes[classId] ":" label;
}
//Display the label at the top of the bounding box
int baseLine;
Size labelSize = getTextSize(label, FONT_HERSHEY_SIMPLEX, 0.8, 1, &baseLine);
top = max(top, labelSize.height);
putText(frame, label, Point(left, top), FONT_HERSHEY_SIMPLEX, 0.8, Scalar(0, 255, 0), 2);
}
// Remove the bounding boxes with low confidence using non-maxima suppression
void postprocess(Mat& frame, const vector<Mat>& outs)
{
vector<int> classIds;
vector<float> confidences;
vector<Rect> boxes;
for (size_t i = 0; i < outs.size(); i)
{
// Scan through all the bounding boxes output from the network and keep only the
// ones with high confidence scores. Assign the box's class label as the class
// with the highest score for the box.
float* data = (float*)outs[i].data;
for (int j = 0; j < outs[i].rows; j, data = outs[i].cols)
{
Mat scores = outs[i].row(j).colRange(5, outs[i].cols);
Point classIdPoint;
double confidence;
// Get the value and location of the maximum score
minMaxLoc(scores, 0, &confidence, 0, &classIdPoint);
if (confidence > confThreshold)
{
int centerX = (int)(data[0] * frame.cols);
int centerY = (int)(data[1] * frame.rows);
int width = (int)(data[2] * frame.cols);
int height = (int)(data[3] * frame.rows);
int left = centerX - width / 2;
int top = centerY - height / 2;
classIds.push_back(classIdPoint.x);
confidences.push_back((float)confidence);
boxes.push_back(Rect(left, top, width, height));
}
}
}
// Perform non maximum suppression to eliminate redundant overlapping boxes with
// lower confidences
vector<int> indices;
NMSBoxes(boxes, confidences, confThreshold, nmsThreshold, indices);
for (size_t i = 0; i < indices.size(); i)
{
int idx = indices[i];
Rect box = boxes[idx];
drawPred(classIds[idx], confidences[idx], box.x, box.y,
box.x box.width, box.y box.height, frame);
}
}
int main()
{
Mat img = imread("./1.jpg");
if (img.empty())
{
cout << "Image read error, please check again!" << endl;
}
string line;
while (getline(ifs, line))
{
classes.push_back(line);
}
net.setPreferableBackend(DNN_BACKEND_OPENCV);
net.setPreferableTarget(DNN_TARGET_CPU);
// Create a 4D blob from a frame.
Mat blob;
blobFromImage(img, blob, 1 / 255.0, Size(inpWidth, inpHeight), Scalar(0, 0, 0), true, false);
//Sets the input to the network
net.setInput(blob);
// Runs the forward pass to get output of the output layers
vector<Mat> outs;
net.forward(outs, getOutputsNames(net));
// Remove the bounding boxes with low confidence
postprocess(img, outs);
// Put efficiency information. The function getPerfProfile returns the
// overall time for inference(t) and the timings for each of the layers(in layersTimes)
vector<double> layersTimes;
double freq = getTickFrequency() / 1000;
double t = net.getPerfProfile(layersTimes) / freq;
string label = format("Inference time for a frame : %.2f ms", t);
putText(img, label, Point(0, 20), FONT_HERSHEY_SIMPLEX, 0.8, Scalar(255, 255, 0), 2);
namedWindow("OpenCV_YoloV4_Demo", WINDOW_NORMAL);
imshow("OpenCV_YoloV4_Demo", img);
waitKey(0);
return 0;
}
下面是OpenCV DNN读取YoloV4模型进行图片检测代码和效果演示：
代码语言：javascript复制int main()
{
string line;
while (getline(ifs, line))
{
classes.push_back(line);
}
VideoCapture cap("./cars.mp4");
Mat frame;
while (1)
{
if (!cap.isOpened())
{
cout << "Video open failed, please check!" << endl;
break;
}
cap.read(frame);
if (frame.empty())
{
cout << "frame is empty, please check!" << endl;
break;
}
net.setPreferableBackend(DNN_BACKEND_OPENCV);
net.setPreferableTarget(DNN_TARGET_CPU);
// Create a 4D blob from a frame.
Mat blob;
blobFromImage(frame, blob, 1 / 255.0, Size(inpWidth, inpHeight), Scalar(0, 0, 0), true, false);
//Sets the input to the network
net.setInput(blob);
// Runs the forward pass to get output of the output layers
vector<Mat> outs;
net.forward(outs, getOutputsNames(net));
// Remove the bounding boxes with low confidence
postprocess(frame, outs);
// Put efficiency information. The function getPerfProfile returns the
// overall time for inference(t) and the timings for each of the layers(in layersTimes)
vector<double> layersTimes;
double freq = getTickFrequency() / 1000;
double t = net.getPerfProfile(layersTimes) / freq;
string label = format("Inference time for a frame : %.2f ms", t);
putText(frame, label, Point(0, 20), FONT_HERSHEY_SIMPLEX, 0.8, Scalar(255, 255, 0), 2);
namedWindow("OpenCV_YoloV4_Demo", WINDOW_NORMAL);
imshow("OpenCV_YoloV4_Demo", frame);
int c = waitKey(1);
if (c == 27)
break;
}
return 0;
}
使用net.setPreferableBackend()函数还可以设置OpenVINO加速，后续有机会再做介绍，谢谢！
