Image Matching Algorithm for Remote Sensing based on FAST-9 and SURF

A novel remote sensing image matching algorithm based on FAST-9 and SURF is proposed. Firstly, the FAST-9 is employed to detect feature points for remote sensing images, and the ID3 algorithm and non maximum method are used to determine the correct corners. And then, main orientation is assigned by the circular region around the interest point. The SURF descriptor is constructed by the square region aligned to the selected orientation. Finally, the matching pairs are calculated by the nearest neighbor distance matching method. Experimental results show that the proposed matching algorithm is effective for remote sensing images, which has excellent invariance.