Harris Corner Detection Algorithm

This article discusses the Harris corner detection algorithm, where the number of detected corners can be controlled by modifying the values of sigma, thresh, radius, and disp parameters. The algorithm works by computing the autocorrelation matrix using image gradients and analyzing its eigenvalues to distinguish corners from edges and flat regions. Additionally, adjusting these parameters can alter the detection accuracy and stability of corner points. For implementation, different image preprocessing methods can be incorporated to increase corner detection quantity, such as blurring (using Gaussian filters), sharpening (via convolution kernels), and edge detection (employing operators like Sobel or Canny). Furthermore, various corner detection algorithms (like Shi-Tomasi or FAST) can be implemented alongside Harris for comparative analysis to seek better results. The Harris algorithm typically involves calculating the corner response function R = det(M) - k(trace(M))^2, where M is the structure tensor matrix. In summary, by systematically adjusting these parameters and exploring different methodological approaches, one can optimize corner detection performance through iterative experimentation and algorithm tuning.