Uav Remote Sensing Image Stitching Based On Point Features

With the development of UAV technology,UAV are widely used in various aerial and remote sensing scenarios.However,the coverage of single remote sensing images of UAV is relatively small.It cannot meet the actual application requirements.How to stitch multiple remote sensing images into a seamless natural image has become a hot topic in recent years.This paper focuses on the strong demand for image stitching in the field of remote sensing,and deeply studies the remote sensing image stitching technology of UAV based on point features.The specific research content is as follows:1.According to the requirements of image collection,this project has independently built a UAV image acquisition system,and has independently compiled the UAV ground control software based on the needs of the project.2.Analyze the Harris corners,SURF features and SIFT features,compare the performance of the three features,and then improve the classic SIFT algorithm,perform distance constraints,maximum number constraints,and minimum response value constraints on the extracted SIFT features.Then,A comparison experiment was conducted between the improved SIFT algorithm and the classic SIFT algorithm.Experimental results show that the improved SIFT algorithm has higher registration accuracy and faster calculation speed,while ensuring the accuracy of image alignment.3.For the problems of uneven brightness,distortion and ghosting in the process of image fusion,this paper proposes to use a bundling adjustment algorithm to globally adjust the multiple images after registration;Next,use the best suture search algorithm based on graph cut to find the best suture,and finally use a multi-band fusion algorithm to fuse the images near the optimal suture.At the same time,the image fusion algorithm proposed in this paper is compared with the classical image fusion algorithm.The experimental results show that the proposed image fusion method has a better fusion effect.