gino0717's profile - activity

can I use yolo v4 model in dnn in opencl? I'm doing some experiment to benchmark the speed of different backend of yolo

WINDOW_NORMAL worked weird after uncheck WITH_QT hi, not sure whether Qt has something to do with this, I took a look a

you may need some grabcut or gmm method (which is good for the image with similar color in ROI but bad contours) to ext

no sure whether the problem or not but your width and height look inverse.

I find some good example of this algorithm. in practice, I found that Fourier based template matching works not so well

trying to disable VIDEOIO ERROR message I'm making a sample function to detect whether the webcam is plugged in the USB

so I always use c++ to make such program

add include(opencv.pri) to your Test.pro

Not sure but you may try some following step: when using the "ldconfig -v" , you may see something like: libopencv_co

Not sure but you may try some following step: when using the "ldconfig -v" , you may see something like: libopencv_co

Not sure but you may try some following step: when using the "ldconfig -v" , you may see something like: libopencv_co

Not sure but you may try some following step: when using the "ldconfig -v" , you may see something like: libopencv_co

Not sure but you may try some following step: when using the "ldconfig -v" , you may see something like: libopencv_co

Not sure but you may try some following step: 1. when using the "ldconfig -v" , you may see something like: libopen

Not sure but you may try some following step: 1. when using the "ldconfig -v" , you may see something like: libopen

Not sure but you may try some following step: 1. when using the "ldconfig -v" , you may see something like: libopen

Not sure but you may try some following step: 1. when using the "ldconfig -v" , you may see something like: libopen

Not sure but you may try some following step: 1. when using the "ldconfig -v" , you may see something like: libo

I guess you are not setting the INCLUDEPATH and LIBS in your *.pro file in QT appropriate.

I guess the version 3.1 is hardcode in your makefile or Cmake file.

Maybe like this? Rect roiRegion=Rect(x,y,x1-x,y1-y); Mat myRoiMat=src(roiRegion); DoSomethingToGetFeature(myRoiMat);

Maybe like this? Rect roiRegion=Rect(x,y,x1-x,y1-y); Mat myRoiMat=src(roiRegion); DoSomethingToGetFeature(myRoiMat);

like this? uchar* yellow_array=src.ptr<uchar>(0); uchar* orange_array=src.ptr<uchar>(1); uchar* green_array

like this? uchar* yellow_array=src.ptr<uchar>(0); uchar* orange_array=src.ptr<uchar>(1); uchar* green_array

like this? uchar* yellow_array=src.ptr<uchar>(0); uchar* orange_array=src.ptr<uchar>(1); uchar*

I always think a white object is difficult to track for some reason: 1. white balance issue: the objects which we human

try swap your channel 0 and channel 2 like: Scalar yellowRgbMin = new Scalar(66, 127, 139); Scalar yellowRgbMax = new S

try swap your channel 0 and channel 2 like: Scalar yellowRgbMin = new Scalar(66, 127, 139); Scalar yellowRgbMax = new Sc

the upper limit of H would be 179 but your H in yellowRgbMax is 249, which may make more colors(especially red) show in

full body tracking is a difficult quest. You may find some solution in openvino or take a look at openpose.

as I know CascadeClassifier in opencv is not so fast. You may try some dlib face detection.

in your project, go to [Project(P)] -> [ (your project name) property ] -> [VC++ Directories] you would see some

not sure but it seems like some decimal point issue here. For example: A=133 B=51 A/2= 133/2=67 ( no decimal points are

in your project, go to [Project(P)] -> [ (your project name) property ] -> [VC++ Directories] you would see some

thanks for the answer. But actually what I need is to detect the yellow color.

color detection in different background of human machine interface Now I want to use a camera to help me look at the hum

CLAHE has obvious grid artifacts I get a camera with a gray surrounding circular shadow and it affects the result of my

I always like to draw the contour in some Mat and use the Hough Line Transform to find lines in that Mat. Then I can use

I'm considering a condition: if my computer is running but the USB of a webcam is accidentally unplugged. I want to write something to make the webcam can re-initialize.

int main()
{
   VideoCapture video(0);
    Mat frame;
   while (true) {
       if(video.isOpened()&&video.grab())
       {
           video>>frame;
           if(frame.cols>0)
           imshow("show video",frame);
           waitKey(2);
       }
       else {
               video.release();
               cout<<"wait key input"<<endl;
               char temp[1]; //get a key input
               cin>>temp;
               if(temp[0]!=NULL)
               {
                  VideoCapture video_temp(0);
                  video=video_temp;
               }
       }
   }
}

1. not sure but it seems video.isOpened() is not always return false even if the USB webcam is not available any more after the webcam first be launched so I may need a video.grab() to do some double check.

2. in Linux the system can immediately detect the disconnect of webcam. However, in windows system, it takes a while ( about 30 seconds ) to find out that the webcam is disconnected.

is there any better way to make the webcam re-initialize in the program?

better ways to re-initialize webcam via VideoCapture? I'm considering a condition: if my computer is running but the USB

Also, notice that the H of red color in HSV color space may range from 0~10 and 170~180, you might have to consider thes

I guess you used GPU in darknet but CPU in openCV dnn

don't know if it work or not, I just thinking if you can use some template of points from a single pill and match it one

I always think CSRT is a little too slow (and it has no intrinsic boundary checking so if the tracking box is near the b

really interesting topic. I'm thinking if some iterative closest point methods can handle this or not.

I find recently people in the forum like to use tesseract. Why would people tend to choose this method these days?