Evaluation of Objects Recognition Efficiency on Mapes by Various Methods

摘要

The paper analyzes the efficiency of image recognition on terrestrial photographs by SURF, SIFT and ORB methods. It has been shown that for high-quality images, the highest probability of recognition in the application of the SIFT method. In the case of identifying fragments of images on noisy and blurred images, the best results are obtained using the ORB method, which, together with this, has the highest performance among the methods used.