Selection of Singular Points Stable to Blurring and Geometric Distortions of Images
A. K. Shakenov · Optoelectronics Instrumentation and Data Processing · 2021
Abstract The method for selecting the stable singular points of images obtained with BRISK and AKAZE algorithms is proposed. The method is based on separation of points preserved on the image after its considerable distortion. The results of numerical experiments on the basis of 1000 images are given confirming the efficiency of the proposed method. It is established that the sets of points obtained with the proposed method contain 3–5 times more stable points than random sets of the same size. It is shown that, for selecting the stable points, it is more efficient to use considerable distortions of images.