Regarding Issue in undistorting RGB Image

C++
,
1

RGB image undistortion was performed using BORDER_CONSTANT and interpolation (INTER_LINEAR). We utilized the initUndistortRectifyMap and remap functions in OpenCV version 4.2.0 and 4.10.0. The undistorted image is expected to have 0 values for out-of-bounds pixels, but the image does not display 0s. Why is this?

2

show us the input image, the result image, and the relevant code, including any third party library calls for “displaying” your result image.

3

Thanks crackwitz for early reply.

We have attached images below.
rgb_image


#include <opencv2/opencv.hpp>
#include <opencv2/imgproc.hpp>
#include <iostream>

using namespace cv;

int main()
{
    // Load the image
    cv::Mat rgb_image = cv::imread("rgb_image.png", cv::IMREAD_UNCHANGED); 

    // Get the number of channels in the image
    int channels = rgb_image.channels(); 
   // Print the number of channels
    std::cout << "Number of channels: " << channels << std::endl;

    // RGB Camera intrinsic parameters (you need to replace these with your actual parameters)
    Mat camera_Matrix_rgb = (Mat_<double>(3, 3) << 704.642972, 0, 318.955605,
                             0, 704.345741, 241.220241,
                             0, 0, 1);                                                                // Intrinsic parameters: fx, fy, cx, cy
    Mat distCoeffs_rgb = (Mat_<double>(1, 5) << 0.007755, -0.060623, -0.000348, -0.000087, 0.000000); // Distortion coefficients: k1, k2, p1, p2, k3

    // Define width and height of the rgb  and depth image
    int width_rgb = 640;  // example width
    int height_rgb = 480; // example height

    // Undistort the rgb image / Generate the rectification map

    cv::Mat map_x_rgb, map_y_rgb;
    initUndistortRectifyMap(camera_Matrix_rgb, distCoeffs_rgb, Mat(), camera_Matrix_rgb, Size(width_rgb, height_rgb), CV_32FC1, map_x_rgb, map_y_rgb);

    // Remap the rgb image to rectify it
    cv::Mat rectified_rgb_image;
    remap(rgb_image, rectified_rgb_image, map_x_rgb, map_y_rgb, INTER_LINEAR, cv::BORDER_CONSTANT, Scalar(0, 0, 0));

    // Save the cropped image
    cv::imshow("rectified_rgb_image", rectified_rgb_image);
    cv::imwrite("rectified_rgb_image.png", rectified_rgb_image);

    cv::waitKey(0);
    cv::destroyAllWindows();

    return 0;
}

We are not able to attach the rectified_rgb_image (result image).
You will get the same from the mentioned code.