A matching algorithm based on improved SIFT for the air mobility platform

The matching of the images, which obtained by the air mobility platform, containing a large amount of image information, requires better real-time performance. The SIFT descriptor vector will be improved in this paper. In addition, preliminary to the matching, the image segmentation will be done by using the Tsallis entropy method. The experimental results show that the proposed algorithm has greatly improved the efficiency without too much damage to the premise of the matching results, and better able to meet the real-time requirements of the processing to the images obtained by the air mobility platform.