opencv进阶-检测使用鼠标框选的自定义区域

API-setMouseCallback

鼠标响应处理函数
void setMousecallback(const string& winname, MouseCallback onMouse, void* userdata=0)

  1. winname:窗口的名字
  2. onMouse:鼠标响应函数,回调函数。指定窗口里每次鼠标时间发生的时候,被调用的函数指针。
    这个函数的原型应该为void on_Mouse(int event, int x, int y, int flags, void* param);
    onMouse的函数原型:
    void on_Mouse(int event, int x, int y, int flags, void* param);
    event是 CV_EVENT_*变量之一
    x和y是鼠标指针在图像坐标系的坐标(不是窗口坐标系)
    flags是CV_EVENT_FLAG的组合, param是用户定义的传递到setMouseCallback函数调用的参数。
  3. userdate:传给回调函数的参数

event鼠标事件总共有10中,从0-9依次代表如下:

#define CV_EVENT_MOUSEMOVE 0 滑动
#define CV_EVENT_LBUTTONDOWN 1 左键点击
#define CV_EVENT_RBUTTONDOWN 2 右键点击
#define CV_EVENT_MBUTTONDOWN 3 中间点击
#define CV_EVENT_LBUTTONUP 4 左键释放
#define CV_EVENT_RBUTTONUP 5 右键释放
#define CV_EVENT_MBUTTONUP 6 中间释放
#define CV_EVENT_LBUTTONDBLCLK 7 左键双击
#define CV_EVENT_RBUTTONDBLCLK 8 右键双击
#define CV_EVENT_MBUTTONDBLCLK 9 中间释放

一、选区矩形区域实时显示相机视频

代码

//---------------------------------【头文件、命名空间包含部分】-----------------------------
//		描述:包含程序所使用的头文件和命名空间
//-------------------------------------------------------------------------------------------------
#include <opencv2/opencv.hpp>
#include <iostream>
using namespace cv;
using namespace std;
//-----------------------------------【宏定义部分】--------------------------------------------
//  描述:定义一些辅助宏 
//------------------------------------------------------------------------------------------------ 
#define WINDOW_NAME "【程序窗口】"        //为窗口标题定义的宏 


//-----------------------------------【全局函数声明部分】------------------------------------
//		描述:全局函数的声明
//------------------------------------------------------------------------------------------------
void on_MouseHandle(int event, int x, int y, int flags, void* param);    //鼠标回调函数

																		 //-----------------------------------【全局变量声明部分】-----------------------------------
																		 //		描述:全局变量的声明
																		 //-----------------------------------------------------------------------------------------------
Rect select;
bool select_flag = false, flag = true;
Point origin;
Mat org, dst, gray_image;

//-----------------------------------main( )函数】--------------------------------------------
//		描述:控制台应用程序的入口函数,我们的程序从这里开始执行
//-------------------------------------------------------------------------------------------------
int main(int argc, char** argv)
{

	VideoCapture capture(0);
	capture >> org;

	select = Rect(-1, -1, 0, 0);
	namedWindow(WINDOW_NAME);//定义一个img窗口  
	namedWindow("dst");
	setMouseCallback(WINDOW_NAME, on_MouseHandle, 0);//调用回调函数 
	while (1)
	{
		if (!capture.read(org))   //获取视频帧失败
		{
			cout << "Cannot read the frame from video file" << endl;
			break;
		}
		resize(org, org, Size(org.cols / 2, org.rows / 2), (0, 0), (0, 0), 3);
		if (flag)
		{
			flag = false;
			select = Rect(10, org.rows / 6, org.cols - 20, org.rows * 3 / 4);//第一次进入循环,记录起始点
			cout << select.x << "--" << select.y << "--" << select.width << "--" << select.height << endl;
		}
		rectangle(org, select, Scalar(255, 0, 0), 1, 8, 0);//能够实时显示在画矩形窗口时的痕迹  
		dst = org(Rect(select.x, select.y, select.width, select.height)); //将感兴趣区域复制到tmp1   
																		  //img = dst.clone();
		cvtColor(dst, gray_image, COLOR_BGR2GRAY);  //彩色图片转换成黑白图片
												 //addWeighted(dst,0.1,img,0.7,0.,dst);
		select = Rect(select.x + 10, select.y + 10, select.width - 20, select.height - 20);//记录起始点
		rectangle(org, select, Scalar(255, 0, 0), 1, 8, 0);//能够实时显示在画矩形窗口时的痕迹  
		select = Rect(select.x - 10, select.y - 10, select.width + 20, select.height + 20);//记录起始点

		rectangle(gray_image, Rect(10, 10, select.width - 20, select.height - 20), Scalar(255, 0, 0), 1, 8, 0);//能够实时显示在画矩形窗口时的痕迹  
		cout << select.x << "--" << select.y << "--" << select.width << "--" << select.height << endl;

		imshow(WINDOW_NAME, org);
		imshow("dst", gray_image);
		if (waitKey(10) == 27) break;//按下ESC键,程序退出
	}
	waitKey(0);
	destroyAllWindows();
	return 0;
}

//--------------------------------on_MouseHandle( )函数】-----------------------------
//		描述:鼠标回调函数,根据不同的鼠标事件进行不同的操作
//--------------------------------------------------------------------------------------
void on_MouseHandle(int event, int x, int y, int, void* param)
{
	//Point origin;//不能在这个地方进行定义,因为这是基于消息响应的函数,执行完后origin就释放了,所以达不到效果。  
	if (select_flag)
	{
		select.x = MIN(origin.x, x);//不一定要等鼠标弹起才计算矩形框,而应该在鼠标按下开始到弹起这段时间实时计算所选矩形框  
		select.y = MIN(origin.y, y);
		select.width = abs(x - origin.x);//算矩形宽度和高度  
		select.height = abs(y - origin.y);
		select &= Rect(0, 0, org.cols, org.rows);//保证所选矩形框在视频显示区域之内  
	}
	if (event == EVENT_LBUTTONDOWN)
	{
		select_flag = true;//鼠标按下的标志赋真值  
		origin = Point(x, y);//保存下来单击是捕捉到的点  
		select = Rect(x, y, 0, 0);//这里一定要初始化,宽和高为(0,0)是因为在opencvRect矩形框类内的点是包含左上角那个点的,但是不含右下角那个点 
	}
	else if (event == EVENT_LBUTTONUP)
	{
		select_flag = false;
	}
}

二、基于SSD检测特定区域内的多种物体

#include 
#include 
#include 

using namespace cv;
using namespace cv::dnn;
using namespace std;

const size_t width = 300;
const size_t height = 300;
const float meanVal = 127.5;
const float scaleFactor = 0.007843f;
const char* classNames[] = { "background",
"aeroplane", "bicycle", "bird", "boat",
"bottle", "bus", "car", "cat", "chair",
"cow", "diningtable", "dog", "horse",
"motorbike", "person", "pottedplant",
"sheep", "sofa", "train", "tvmonitor" };

String labelFile = "D:/opencv-4.4.0/models/ssd/labelmap_det.txt";
String model_text_file = "D:/opencv-4.4.0/models/ssd/MobileNetSSD_deploy.prototxt";
String modelFile = "D:/opencv-4.4.0/models/ssd/MobileNetSSD_deploy.caffemodel";

#define WINDOW_NAME "【程序窗口】" 

void on_MouseHandle(int event, int x, int y, int flags, void* param);
Rect select;
bool select_flag = false, flag = true;
Point origin;
Mat org, frame, gray_image, binary_image;

int main(int argc, char** argv)
{
	select = Rect(-1, -1, 0, 0);
	namedWindow(WINDOW_NAME);//定义一个img窗口  
	namedWindow("dst");
	setMouseCallback(WINDOW_NAME, on_MouseHandle, 0);//调用回调函数
 
	VideoCapture capture(0);
	capture >> org;
	Net net = readNetFromCaffe(model_text_file, modelFile);
	while (1)
	{
		if (!capture.read(org))   //获取视频帧失败
		{
			cout << "Cannot read the frame from video file" << endl;
			break;
		}

		if (flag)
		{
			flag = false;
			select = Rect(10, org.rows / 6, org.cols - 20, org.rows * 3 / 4);//第一次进入循环,记录起始点
																			 //cout << select.x << "--" << select.y << "--" << select.width << "--" << select.height << endl;
		}
		rectangle(org, select, Scalar(255, 0, 0), 1, 8, 0);//能够实时显示在画矩形窗口时的痕迹  
		frame = org(Rect(select.x, select.y, select.width, select.height)); //将感兴趣区域复制到tmp1   

		Mat inputblob = blobFromImage(frame, scaleFactor, Size(width, height), meanVal, false);
		net.setInput(inputblob, "data");
		net.setPreferableBackend(DNN_BACKEND_OPENCV);
		net.setPreferableTarget(DNN_TARGET_CPU);
		Mat detection = net.forward("detection_out");

		Mat detectionMat(detection.size[2], detection.size[3], CV_32F, detection.ptr<float>());
		float confidence_threshold = 0.25;
		for (int i = 0; i < detectionMat.rows; i++) {
			float confidence = detectionMat.at<float>(i, 2);
			if (confidence > confidence_threshold) {
				size_t objIndex = (size_t)(detectionMat.at<float>(i, 1));
				float tl_x = detectionMat.at<float>(i, 3) * frame.cols;
				float tl_y = detectionMat.at<float>(i, 4) * frame.rows;
				float br_x = detectionMat.at<float>(i, 5) * frame.cols;
				float br_y = detectionMat.at<float>(i, 6) * frame.rows;

				Rect object_box((int)tl_x, (int)tl_y, (int)(br_x - tl_x), (int)(br_y - tl_y));
				rectangle(frame, object_box, Scalar(0, 0, 255), 2, 8, 0);
				printf("%s:%.2f\n", classNames[objIndex],  confidence);//打印出检测的消息
				putText(frame, format("%s:%.2f,%d", classNames[objIndex], confidence, i + 1), Point(tl_x, tl_y + 20), FONT_HERSHEY_SIMPLEX, 0.7, Scalar(255, 255, 0), 2);//并且在输出的图像上打上标记
			}
		}
		vector < double>layerstimings;
		double freq = getTickFrequency() / 1000;
		double time = net.getPerfProfile(layerstimings) / freq;
		ostringstream ss;
		//ss << "FPS" << 1000 / time << ";time:" << time << "ms";
		//putText(frame, ss.str(), Point(20, 20), FONT_HERSHEY_PLAIN, 1, Scalar(0, 0, 255), 2, 8);

		//select = Rect(select.x + 10, select.y + 10, select.width - 20, select.height - 20);//记录起始点
		//rectangle(org, select, Scalar(255, 0, 0), 1, 8, 0);//能够实时显示在画矩形窗口时的痕迹

		//select = Rect(select.x - 10, select.y - 10, select.width + 20, select.height + 20);//记录起始点
		//rectangle(frame, Rect(10, 10, select.width - 20, select.height - 20), Scalar(255, 0, 0), 1, 8, 0);//能够实时显示在画矩形窗口时的痕迹  
		//cout << select.x << "--" << select.y << "--" << select.width << "--" << select.height << endl;

		imshow(WINDOW_NAME, org);
		imshow("dst", frame);
		if (waitKey(10) == 27) break;//按下ESC键,程序退出
		waitKey(200);//窗口切换延迟需要满足实时检测的要求,这个时间要大于读取每一帧的时间
	}
	waitKey(0);
	destroyAllWindows();
	return 0;
}

void on_MouseHandle(int event, int x, int y, int, void* param)
{
	if (select_flag)
	{
		select.x = MIN(origin.x, x);//不一定要等鼠标弹起才计算矩形框,而应该在鼠标按下开始到弹起这段时间实时计算所选矩形框  
		select.y = MIN(origin.y, y);
		select.width = abs(x - origin.x);//算矩形宽度和高度  
		select.height = abs(y - origin.y);
		select &= Rect(0, 0, org.cols, org.rows);//保证所选矩形框在视频显示区域之内  
	}
	if (event == EVENT_LBUTTONDOWN)
	{
		select_flag = true;//鼠标按下的标志赋真值  
		origin = Point(x, y);//保存下来单击是捕捉到的点  
		select = Rect(x, y, 0, 0);//这里一定要初始化,宽和高为(0,0)是因为在opencv中Rect矩形框类内的点是包含左上角那个点的,但是不含右下角那个点 
	}
	else if (event == EVENT_LBUTTONUP)
	{
		select_flag = false;
	}
}


YOLOV4自定义区域检测

#include 
#include 
#include 
#include 
#include 
#include 

using namespace std;
using namespace cv;
using namespace cv::dnn;

#define WINDOW_NAME "【程序窗口】"       

void on_MouseHandle(int event, int x, int y, int flags, void* param);    //鼠标回调函数

Rect select;
bool select_flag = false, flag = true;
Point origin;
Mat org, frame;

String yolo_cfg = "D:/opencv-4.4.0/models/yolov4/yolov4-voc-ball-test.cfg";
String yolo_model = "D:/opencv-4.4.0/models/yolov4/yolov4-voc-BOLT_3000.weights";

int main()
{
	Net net = readNetFromDarknet(yolo_cfg, yolo_model);
	vector<string> classNamesVec;
	ifstream classNamesFile("D:/opencv-4.4.0/models/yolov4/bolt.names");
	if (classNamesFile.is_open())
	{
		string className = "";
		while (std::getline(classNamesFile, className))
			classNamesVec.push_back(className);
	}

	//net.setPreferableBackend(DNN_BACKEND_INFERENCE_ENGINE);
	net.setPreferableTarget(DNN_TARGET_CPU);
	net.setPreferableBackend(DNN_BACKEND_OPENCV);

	std::vector<String> outNames = net.getUnconnectedOutLayersNames();
	for (int i = 0; i < outNames.size(); i++) {
		printf("output layer name : %s\n", outNames[i].c_str());
	}

	namedWindow(WINDOW_NAME);//定义一个img窗口  
	namedWindow("dst");
	setMouseCallback(WINDOW_NAME, on_MouseHandle, 0);//调用回调函数
	VideoCapture capture("D:/images/IMG_4753.mp4");

	while (1)
	{
		capture >> org;//读取的视频寄存在org画布内
		if (!capture.read(org))   //获取视频帧失败
		{
			cout << "Cannot read the frame from video file" << endl;
			break;
		}
		//将org画图缩放大小
		resize(org, org, Size(0, 0), 0.2, 0.2, 3);
		if (flag)
		{
			flag = false;
			select = Rect(20, 20, org.cols / 2, org.rows / 2);//第一个框的大小
		}
		rectangle(org, select, Scalar(255, 0, 0), 1, 8, 0); //显示第一个框
		frame = org(Rect(select.x, select.y, select.width, select.height)); //将感兴趣区域复制到dst  
		Mat inputBlob = blobFromImage(frame, 1 / 255.F, Size(416, 416), Scalar(), true, false);
		net.setInput(inputBlob);

		// 输出检测频率和每帧耗时
		std::vector<Mat> outs;
		net.forward(outs, outNames);
		vector<double> layersTimings;
		double freq = getTickFrequency() / 1000;
		double time = net.getPerfProfile(layersTimings) / freq;
		ostringstream ss;
		ss << "FPS" << 1000 / time << ";time:" << time << "ms";
		putText(frame, ss.str(), Point(20, 20), FONT_HERSHEY_PLAIN, 1, Scalar(0, 0, 255), 2, 8);

		// 输出检测框和置信度
		vector<Rect> boxes;
		vector<int> classIds;
		vector<float> confidences;
		for (size_t i = 0; i < outs.size(); ++i)
		{
			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;
				minMaxLoc(scores, 0, &confidence, 0, &classIdPoint);
				if (confidence > 0.3)
				{
					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));

				}
			}
		}

		vector<int> indices;
		NMSBoxes(boxes, confidences, 0.5, 0.2, indices);
		for (size_t i = 0; i < indices.size(); ++i)
		{
			int idx = indices[i];
			Rect box = boxes[idx];
			String className = classNamesVec[classIds[idx]];
			putText(frame, format("%s:%.2f ", className.c_str(), confidences[idx]), box.tl(), FONT_HERSHEY_SIMPLEX, 0.5, Scalar(255, 0, 0), 2, 8);
			rectangle(frame, box, Scalar(0, 0, 255), 2, 8, 0);
		}
		
		imshow(WINDOW_NAME, org);
		imshow("dst", frame);
		char c = waitKey(5);
		if (c == 27) { // ESC退出
			break;
		}
		waitKey(200);
	}
	capture.release();//释放资源
	waitKey(0);
	return 0;
}

void on_MouseHandle(int event, int x, int y, int, void* param)
{
	//Point origin;//不能在这个地方进行定义,因为这是基于消息响应的函数,执行完后origin就释放了,所以达不到效果。  
	if (select_flag)
	{
		select.x = MIN(origin.x, x);//不一定要等鼠标弹起才计算矩形框,而应该在鼠标按下开始到弹起这段时间实时计算所选矩形框  
		select.y = MIN(origin.y, y);
		select.width = abs(x - origin.x);//算矩形宽度和高度  
		select.height = abs(y - origin.y);
		select &= Rect(0, 0, org.cols, org.rows);//保证所选矩形框在视频显示区域之内  
	}

	if (event == EVENT_LBUTTONDOWN)
	{
		select_flag = true;//鼠标按下的标志赋真值  
		origin = Point(x, y);//保存下来单击是捕捉到的点  
		select = Rect(x, y, 0, 0);//这里一定要初始化,宽和高为(0,0)是因为在opencv中Rect矩形框类内的点是包含左上角那个点的,但是不含右下角那个点 
	}

	else if (event == EVENT_LBUTTONUP)
	{
		select_flag = false;
	}
}

问题:对比全局检测,自定义区域检测的速度有无不同?

就SSD检测物体而言,速度有所减低,从原来的11帧每秒降低至9帧每秒。

reference
OPENCV2 中视频处理——用鼠标选定矩形框并实时处理显示
opencv 鼠标拖拽选择感兴趣区域

你可能感兴趣的:(opencv,机器学习)