Image Stitching Method Based on Adaptive Weighted Fusion
Chunbo Xiu, Jingyao Fang, Jiang Zhang · 2021
Aiming at the problem of stitching seams and uneven brightness caused by illumination and noise during the image stitching process, a stitching method based on adaptive weighted fusion is proposed. The SURF algorithm is used to extract the feature points of the image to be stitched, the Euclidean distance is used to calculate the similarity of the feature point pairs, and complete the image registration. After that, the adaptive weighted method is applied to fuse the brightness difference of the overlapping area to achieve the smooth transition of the spliced image and eliminate stitching seams. Compared with the traditional gradual algorithm and the nonlinear fusion algorithm, the experiment results show that the adaptive fusion algorithm can obtain better image fusion effect and effectively improve the quality of image stitching.