Aiming at the disadvantage of traditional image mosaic algorithms, an advanced mosaic algorithm based on improved SIFT algorithm was proposed and applied in the stitching of UAV remote sensing image. Firstly, the feature points of the sensing images were extracted using Harris corner detector, then the feature points were described by the improved SIFT algorithm, and the calculation amount was decreased by the dimension reduction process. And after matching, RANSAC method was used to eliminate the inaccurate matching. Finally, the image mosaic was realized with the fading-in-and-out weighted average method. The experimental results show that, this method can effectively eliminate the wrong matching, decrease the time complexity and avoid the splicing gap more efficiently.
