Extract corner points using Krawtchouk polynomials

Abstract

The corner has important features and useful information from the objects in the image. The extraction of corner points is most often used in the preprocessing of image analysis. Therefore, it is considered as one of the most important and fundamental stages of image processing. In addition, it plays an essential role in machine vision and image enhancement, image compression, image matching, image tracking, and so on. In this article, corner points are detected in images using polynomials. A discrete orthogonal polynomial is a useful digital image processing tool for extracting object contours in various applications. Therefore, in the corner detection step algorithm, the Krawtchouk polynomial is used in the edge detection step. In the next step, an anisotropic direction derivative is calculated at each pixel in the contour, and finally the removal of the non-maximum suppression and the threshold in each contour is used to find the corners. The results of the experiments show that the proposed method has a good performance compared to other common algorithms.