ARYL518's profile - activity

Use build_sdk.py to build opencv4 for android on Windows Hi, I need to build OpenCV for Android on Windows10. The OpenCV

Use build_sdk.py to build OpenCV for Android Hi, I use command: D:\build-android> python opencv-master\platforms\andr

how to get a outputarrayofarrays Hi, I have a method func(OutputArrayOfArrays points) { vector < vector < Point

It works. Thanks!

My question is how to build OpenCV with java wrapper for android rather than how to use a prebuild OpenCV-android.

build opencv with java wrapper for android Hi, I added some functionality in the chessboard.cpp and I can compile it on

yes, I can add include path for highgui, but which dependents on other include files which depend on others...

I uncommeted it in chessboard.cpp and get the error: cannot open source file "opencv2/highgui.hpp"

build OpenCV4: add preprocessor definition "CV_DETECTORS_CHESSBOARD_DEBUG" Hi, I am building OpenCV4 on Windows using VS

Actually I know where is the source code. I want to know the reference the algorithm is based on.

I am using OpenCV4.4. cv::details::Chessboard::Board.

white angle and black angle in chessboard Hi, I am reading the source code about two classes: Chessboard and Chessboard:

Thank you so much!

Nonlinear solver such as Levenberg- Marquardt Hi, I need to use a nonlinear solver in OpenCV. Does OpenCV support it or

ICP does not require the correspondence. Since I already know the correspondence, maybe some more efficient methods exi

Find rigid 3D transform between two 3D point sets Hi, I have two 3D point sets with known correspondence and I want to f

Thanks. The noisy measurement you mean is the rotation and translation vector estimated by estimatePoseSingleMarkers, ri

Jitter using ArUco marker Hi, I am using ArUco marker to find the pose. The dictionary is DICT_6X6_250 and the physical

combine six ArUco markers into a cuboid marker Hi, I want to use six ArUco markers and each one is attached on one side

Actually, I want to know the underlying technique to do the pose estimation of a cubic marker. It is better if people ca

Pose estimation of a small cube shape marker Hi, I would like to estimate the pose of a needle, looks like a thin cylind

Thanks a lot. I will try it.

Jitter problem of the pose estimated by SolvePnp Hi, I use SolvePnp to estimate the pose of the head. The 2D points come

Pose estimation of a needle Hi, I would like to real-time to estimate the pose of a needle hold by the hand. The needle

Thanks. I need at leas 6 points rather than 4.

App crash when using solvePnP on Android Hi, I use OpenCV on Andoid. I can use it for face detection. But when I use it

Hi, I want to use opencv-4.1.1-android-sdk in my app. My android studio version is 3.3.2 and sdk api is 28. I use Windows. I copied all libs under native/libs to my jniLibs folder. Then I import new module by specify the sdk/java folder and it succeed. However, when I open module setting and add dependence, I got sync error as following:

ERROR: Unable to resolve dependency for ':app@debug/compileClasspath': Could not resolve project :opencvjavasupport. Show Details Affected Modules: app

Any help are very appreciated.

Thanks.

YL

It works. Many thanks.

sync failed when adding opencv as a dependence for my app Hi, I want to use opencv-4.1.1-android-sdk in my app. My andro

Camera calibration with fixed principal point Hi, I would like to use cv::calibrateCamera to do camera calibration. In m

Is it sufficient just using one image to calibrate a camera with good initialization known? Hi, I have a question for ca

Is it sufficient just using one image to calibrate a camera with good initialization known? Hi, I have a question for ca

Hi, I want to use ArUco marker to do AR and I want to know if I can track the ArUco marker in a grays scale OpenCV image?

Thanks.

YL

Track ArUco marker in gray scale OpenCV image? Hi, I want to use ArUco marker to do AR and I want to know if I can trac

Hi, I am looking the implementation of SimpleBlobDetector. It looks the insertions sort is not correct as shown below. I also copy entire SimpleBlobDetector codes at the end.

For the insertion sort, we should compare with the key or switch values, right? The following OpenCV code looks weird.

size_t k = centers[j].size() - 1;
while( k > 0 && centers[j][k].radius < centers[j][k-1].radius )//compare with key
{
    centers[j][k] = centers[j][k-1];//or switch values here
    k--;
}
centers[j][k] = curCenters[i];

Entire blob detection codes:

void SimpleBlobDetectorImpl::detect(InputArray image, std::vector<cv::KeyPoint>& keypoints, InputArray)
{
    //TODO: support mask
    keypoints.clear();
    Mat grayscaleImage;
    if (image.channels() == 3)
        cvtColor(image, grayscaleImage, COLOR_BGR2GRAY);
    else
        grayscaleImage = image.getMat();

    std::vector < std::vector<Center> > centers;
    for (double thresh = params.minThreshold; thresh < params.maxThreshold; thresh += params.thresholdStep)
    {
        Mat binarizedImage;
        threshold(grayscaleImage, binarizedImage, thresh, 255, THRESH_BINARY);

        std::vector < Center > curCenters;
        findBlobs(grayscaleImage, binarizedImage, curCenters);
        std::vector < std::vector<Center> > newCenters;
        for (size_t i = 0; i < curCenters.size(); i++)
        {
            bool isNew = true;
            for (size_t j = 0; j < centers.size(); j++)
            {
                double dist = norm(centers[j][ centers[j].size() / 2 ].location - curCenters[i].location);
                isNew = dist >= params.minDistBetweenBlobs && dist >= centers[j][ centers[j].size() / 2 ].radius && dist >= curCenters[i].radius;
                if (!isNew)
                {
                    centers[j].push_back(curCenters[i]);

                    size_t k = centers[j].size() - 1;
                    while( k > 0 && centers[j][k].radius < centers[j][k-1].radius )
                    {
                        centers[j][k] = centers[j][k-1];//should switch here?
                        k--;
                    }
                    centers[j][k] = curCenters[i];

                    break;
                }
            }
            if (isNew)
                newCenters.push_back(std::vector<Center> (1, curCenters[i]));
        }
        std::copy(newCenters.begin(), newCenters.end(), std::back_inserter(centers));
    }

    for (size_t i = 0; i < centers.size(); i++)
    {
        if (centers[i].size() < params.minRepeatability)
            continue;
        Point2d sumPoint(0, 0);
        double normalizer = 0;
        for (size_t j = 0; j < centers[i].size(); j++)
        {
            sumPoint += centers[i][j].confidence * centers[i][j].location;
            normalizer += centers[i][j].confidence;
        }
        sumPoint *= (1. / normalizer);
        KeyPoint kpt(sumPoint, (float)(centers[i][centers[i].size() / 2].radius) * 2.0f);
        keypoints.push_back(kpt);
    }
}

Hi, I have NV12 (one of YUV) format video and I want to real-time detect some color balls with diameter around 10mm in the video. The ball will somehow change its color in the video due to the environment light and its reflection. I think the first step is to do the threshold. Which color space is suitable for the thresholding? RGB, YUV or HSV?

Many thanks.

YL

Thanks. Would you please let me know how to do ellipse detection for a YUV format video?

Hi, I have an object with four sphere with coordinates (center of the sphere): A: 0.0, 0.0, 0.0 B: 0.0, 0.0, 50.0 C: 0.0, 25.0, 100.0 D: 0.0, -25.0, 135.0

There four spheres have diameter around 1 cm and have a gray color. The capture video has YUV format. To track the pose of the object, I think I need to first detect the center of the four spheres in the video. Any suggestion about a fast sphere detection?

After detect the spheres, I am not sure treating the object as a planar object and use SolvePNP to estimate pose or treating the object as a 3D object and use POSIT to estimate pose.

Thanks a lot.

YL