A Method of Using Genetic Algorithm in Image Stitching
Lingfeng Zhao, Yujie Huang, Minge Jing, Xiaoyang Zeng, Yibo Fan · 2018
Image stitching is an important part of computer vision, and how to do it more efficiently with high quality is a heated topic. In this paper, the authors propose a new method called TMGA for image stitching to get an improved performance in calculating Transform Matrix by using Genetic Algorithm. The proposed TMGA not only counts the number of interior points, but also takes standard error and degree of dispersion into consideration compared the traditional methods. The results demonstrate that the proposed algorithm can gain a high-quality transform matrix and improves the result of the stitching.