opencv+arduino實(shí)現(xiàn)物體點(diǎn)追蹤效果
更新時(shí)間:2018年01月09日 10:25:53 作者:wi162yyxq
這篇文章主要為大家詳細(xì)介紹了opencv+arduino實(shí)現(xiàn)物體點(diǎn)追蹤效果,具有一定的參考價(jià)值,感興趣的小伙伴們可以參考一下
本文所要實(shí)現(xiàn)的結(jié)果是:通過在攝像頭中選擇一個(gè)追蹤點(diǎn),通過pc控制攝像頭的舵機(jī),使這一點(diǎn)始終在圖像的中心。
要點(diǎn):使用光流法在舵機(jī)旋轉(zhuǎn)的同時(shí)進(jìn)行追蹤,若該點(diǎn)運(yùn)動,則攝像頭跟蹤聯(lián)動。
#include<opencv2\opencv.hpp> #include<opencv\cv.h> #include<opencv\highgui.h> #include<math.h> #include<Windows.h> #include<string.h> using namespace std; using namespace cv; #define WINDOW_NAME "【程序窗口】" void on_MouseHandle(int event, int x, int y, int flags, void* param); void DrawRectangle( cv::Mat& img, cv::Rect box ); void tracking(Mat &frame,vector<Point2f> temp); HANDLE hComm; LPCWSTR pStr=L"COM4"; char lpOutbuffer[100]; DWORD dwbyte=100; Mat srcImage,grayImage,tempImage1,tempImage,imageROI,grayprev; int g_maxCornerNumber = 1; double qualityLevel = 0.01; double minDistance = 10; int blockSize = 3; double k = 0.04; vector<Point2f> corners; vector<Point2f> pre_corners; vector<Point2f> counts; vector<uchar> status; vector<float> err; Rect g_rectangle; Rect g_temprectangle; bool g_bDrawingBox = false; int main( int argc, char** argv ) { Mat frame; Mat result; COMSTAT Comstat; DWORD dwError; BOOL bWritestat; hComm=CreateFile(pStr,GENERIC_READ | GENERIC_WRITE,0,0,OPEN_EXISTING, 0,NULL); if (hComm == INVALID_HANDLE_VALUE) { cout<<"FLASE"; return -1; } else { cout<<"TURE"; } DCB dcb; GetCommState(hComm,&dcb); dcb.BaudRate=9600; dcb.ByteSize=8; dcb.Parity=NOPARITY; dcb.StopBits=TWOSTOPBITS; bool set=SetCommState(hComm,&dcb); bool sup=SetupComm(hComm,1024,1024); VideoCapture capture(0); namedWindow( WINDOW_NAME ); setMouseCallback(WINDOW_NAME,on_MouseHandle,(void*)&frame); while(1) { capture >> frame; if(!frame.empty()) { cvtColor(frame,grayImage,CV_RGB2GRAY); if( g_bDrawingBox ) rectangle(frame,g_rectangle.tl(),g_rectangle.br(),Scalar(255,255,255)); if (corners.size()!=0) { bool can=PurgeComm(hComm,PURGE_TXCLEAR); if (corners[0].x>(frame.cols/2+100)) { lpOutbuffer[0]='a'; bool ne=WriteFile(hComm,lpOutbuffer,dwbyte,&dwbyte,NULL); } else if (corners[0].x<(frame.cols/2-100)) { lpOutbuffer[0]='b'; bool ne=WriteFile(hComm,lpOutbuffer,dwbyte,&dwbyte,NULL); } tracking(frame,corners); rectangle(frame,Point(corners[0].x-10,corners[0].y-10),Point(corners[0].x+10,corners[0].y+10),Scalar(255,255,255)); } imshow( WINDOW_NAME, frame ); } else { printf(" --(!) No captured frame -- Break!"); break; } int c = waitKey(50); if( (char)c == 27 ) { break; } } return 0; } void on_MouseHandle(int event, int x, int y, int flags, void* param) { Mat& image = *(cv::Mat*) param; switch( event) { case EVENT_MOUSEMOVE: { if( g_bDrawingBox ) { g_rectangle.width = x-g_rectangle.x; g_rectangle.height = y-g_rectangle.y; } } break; case EVENT_LBUTTONDOWN: { g_bDrawingBox = true; g_rectangle =Rect( x, y, 0, 0 ); } break; case EVENT_LBUTTONUP: { g_bDrawingBox = false; if( g_rectangle.width < 0 ) { g_rectangle.x += g_rectangle.width; g_rectangle.width *= -1; } if( g_rectangle.height < 0 ) { g_rectangle.y += g_rectangle.height; g_rectangle.height *= -1; } imageROI=grayImage(g_rectangle); goodFeaturesToTrack( imageROI,corners,g_maxCornerNumber,qualityLevel,minDistance,Mat(),blockSize,false,k ); for (int i = 0; i < corners.size(); i++) { corners[i].x=corners[i].x+g_rectangle.x; corners[i].y=corners[i].y+g_rectangle.y; } } break; } } void tracking(Mat &frame,vector<Point2f> temp) { cvtColor(frame, tempImage1, COLOR_BGR2GRAY); if (grayprev.empty()) { tempImage1.copyTo(grayprev); } calcOpticalFlowPyrLK(grayprev, tempImage1, temp, pre_corners, status, err); for (size_t i=0; i<pre_corners.size(); i++) { line(frame, temp[i], pre_corners[i], Scalar(0, 0, 255)); circle(frame, pre_corners[i], 4, Scalar(0, 255, 0), -1,8,0); } swap(pre_corners, corners); swap(grayprev, tempImage1); }
以上就是本文的全部內(nèi)容,希望對大家的學(xué)習(xí)有所幫助,也希望大家多多支持腳本之家。
您可能感興趣的文章:
- Opencv光流運(yùn)動物體追蹤詳解
- opencv3/C++實(shí)現(xiàn)光流點(diǎn)追蹤
- python+opencv實(shí)現(xiàn)動態(tài)物體追蹤
- 使用OpenCV實(shí)現(xiàn)檢測和追蹤車輛
- 如何用OpenCV -python3實(shí)現(xiàn)視頻物體追蹤
- Python+OpenCV實(shí)現(xiàn)實(shí)時(shí)眼動追蹤的示例代碼
- OpenCV3.0+Python3.6實(shí)現(xiàn)特定顏色的物體追蹤
- 淺析Python+OpenCV使用攝像頭追蹤人臉面部血液變化實(shí)現(xiàn)脈搏評估
- OpenCV 顏色追蹤的示例代碼
- 圖文詳解OpenCV中光流以及視頻特征點(diǎn)追蹤
相關(guān)文章
C++如何計(jì)算結(jié)構(gòu)體與對象的大小
這篇文章主要給大家介紹了關(guān)于C++如何計(jì)算結(jié)構(gòu)體與對象大小的相關(guān)資料,文中通過示例代碼介紹的非常詳細(xì),對大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值,需要的朋友們下面隨著小編來一起學(xué)習(xí)學(xué)習(xí)吧2021-05-05C++中sort()函數(shù)和priority_queue容器中比較函數(shù)的區(qū)別詳析
C++中sort()和priority_queue都能自定義比較函數(shù),其中sort()自定義的比較函數(shù)比較好理解,priority_queue中自定義的比較函數(shù)的效果和sort()是相反的,這篇文章主要給大家介紹了關(guān)于C++中sort()函數(shù)和priority_queue容器中比較函數(shù)的區(qū)別的相關(guān)資料,需要的朋友可以參考下2023-03-03c語言函數(shù)棧幀的創(chuàng)建和銷毀過程詳解
我們知道c語言中函數(shù)都是被調(diào)用的,main函數(shù)里面能調(diào)用其他函數(shù),其實(shí)main函數(shù)也是被別的函數(shù)調(diào)用的,下面通過本文給大家分享c語言函數(shù)棧幀的創(chuàng)建和銷毀過程,一起看看吧2021-08-08