A New Approach to Detect Aircrafts in Remote Sensing Images Based on Corner and Edge Information Fusion

This paper presents a new method to detect aircrafts in remote-sensing image, in which the edge and corner features are used.Firstly, the whole image is operated by canny edge detector and then the OTSU binarization is employed to segment it.By these preprocessing procedures, the main information in the image is enhanced.Then Harris corner detection is conducted on the image, finding the areas where the corner quantity meets the standard.Finally the false alarms are removed by a region-growing clustering method.Experiments on 60 remote-sensing images are performed, including 238 airplanes.Totally 220 airplanes are detected, 18 airplanes are missed and 48 false alarms are reported.The experimental results demonstrate that the newly proposed approach can effectively solve aircraft detection problem in complex background, and have a good detection performance.