Структурна класифікація зображень з використанням просторового оброблення описів

Abstract

The problem of structural recognition of visual objects based on descriptions in the form of a set of key points of the image was solved. The problem of object recognition in the image in the presence of local and random noise, geometric transformations, changes in lighting and background, changes in the angle of image capture was solved. An ORB algorithm for finding binary descriptors was proposed. A method for finding the distances between single-bit distributions of input images and reference images for the purpose of their comparison and classification was proposed.