我已经开发了一个程序来使用JavaCV检测运动。到目前为止,我已经完成了处理图像的cvFindContours。源代码如下
public class MotionDetect {
public static void main(String args[]) throws Exception, InterruptedException {
//FFmpegFrameGrabber grabber = new FFmpegFrameGrabber(new File("D:/pool.avi"));
OpenCVFrameGrabber grabber = new OpenCVFrameGrabber("D:/2.avi");
final CanvasFrame canvas = new CanvasFrame("My Image");
final CanvasFrame canvas2 = new CanvasFrame("ROI");
canvas.setDefaultCloseOperation(javax.swing.JFrame.EXIT_ON_CLOSE);
grabber.start();
IplImage frame = grabber.grab();
CvSize imgsize = cvGetSize(frame);
IplImage grayImage = cvCreateImage(imgsize, IPL_DEPTH_8U, 1);
IplImage ROIFrame = cvCreateImage(cvSize((265 - 72), (214 - 148)), IPL_DEPTH_8U, 1);
IplImage colorImage;
IplImage movingAvg = cvCreateImage(imgsize, IPL_DEPTH_32F, 3);
IplImage difference = null;
IplImage temp = null;
IplImage motionHistory = cvCreateImage(imgsize, IPL_DEPTH_8U, 3);
CvRect bndRect = cvRect(0, 0, 0, 0);
CvPoint pt1 = new CvPoint(), pt2 = new CvPoint();
CvFont font = null;
//Capture the movie frame by frame.
int prevX = 0;
int numPeople = 0;
char[] wow = new char[65];
int avgX = 0;
//Indicates whether this is the first time in the loop of frames.
boolean first = true;
//Indicates the contour which was closest to the left boundary before the object
//entered the region between the buildings.
int closestToLeft = 0;
//Same as above, but for the right.
int closestToRight = 320;
while (true) {
colorImage = grabber.grab();
if (colorImage != null) {
if (first) {
difference = cvCloneImage(colorImage);
temp = cvCloneImage(colorImage);
cvConvertScale(colorImage, movingAvg, 1.0, 0.0);
first = false;
//cvShowImage("My Window1", difference);
} //else, make a running average of the motion.
else {
cvRunningAvg(colorImage, movingAvg, 0.020, null);
}
//Convert the scale of the moving average.
cvConvertScale(movingAvg, temp, 1.0, 0.0);
//Minus the current frame from the moving average.
cvAbsDiff(colorImage, temp, difference);
//Convert the image to grayscale.
cvCvtColor(difference, grayImage, CV_RGB2GRAY);
//canvas.showImage(grayImage);
//Convert the image to black and white.
cvThreshold(grayImage, grayImage, 70, 255, CV_THRESH_BINARY);
//Dilate and erode to get people blobs
cvDilate(grayImage, grayImage, null, 18);
cvErode(grayImage, grayImage, null, 10);
canvas.showImage(grayImage);
ROIFrame = cvCloneImage(grayImage);
cvSetImageROI(ROIFrame, cvRect(72, 148, (265 - 72), (214 - 148)));
//cvOr(outFrame, tempFrame, outFrame);
cvShowImage("ROI Frame", ROIFrame);
cvRectangle(colorImage, /* the dest image */
cvPoint(72, 148), /* top left point */
cvPoint(265, 214), /* bottom right point */
cvScalar(255, 0, 0, 0), /* the color; blue */
1, 8, 0);
CvMemStorage storage = cvCreateMemStorage(0);
CvSeq contour = new CvSeq(null);
cvFindContours(grayImage, storage, contour, Loader.sizeof(CvContour.class), CV_RETR_CCOMP, CV_CHAIN_APPROX_SIMPLE);
}
//Show the frame.
cvShowImage("My Window", colorImage);
//Wait for the user to see it.
cvWaitKey(10);
}
//If this is the first time, initialize the images.
//Thread.sleep(50);
}
}
}
在此代码ROIFrame中,我需要计算白色轮廓区域或像素数?? ..有什么我可以继续进行的方法
最佳答案
使用函数cvContourArea()
文档here。
在您的代码中,在cvFindContours之后,对所有轮廓进行循环,如下所示:
CvSeq* curr_contour = contour;
while (curr_contour != NULL) {
area = fabs(cvContourArea(curr_contour,CV_WHOLE_SEQ, 0));
current_contour = current_contour->h_next;
}
不要忘记将区域存储在某个地方。
关于opencv - 使用opencv和Javacv计算轮廓中的白色区域像素,我们在Stack Overflow上找到一个类似的问题:https://stackoverflow.com/questions/13363574/