Modified SURF applied in Remote Sensing Image Stitching

With the development of satellite technology, the application of remote sensing images becomes more and more significant. The coverage of a single remote sensing image is limited, so remote sensing images should be stitched into a larger image by overlapping region. SURF is a commonly algorithm for image mosaic, but the calculation of this algorithm is too heavy. Thus, we propose the method of modified SURF applied in remote sensing image stitching. Firstly, we extract basic feature points by Hessian matrix. Secondly, we improve the accuracy of feature point matches by use of FLANN algorithm and RANSAC. Finally, we take advantage of the algorithm of overlapping region weighted average to blend remote sensing images. Through the experiment, we can prove that the image stitching system can implement remote sensing image mosaic, which enables the smooth transition for image blending and has superiority and efficiency about image stitching.