基于BRISK和改进RANSAC算法的图像拼接  

Image mosaicing based on BRISK and improved RANSAC algorithm

作　　者：杜港 侯凌燕[1] 佟强 杨大利[1] DU Gang;HOU Ling-yan;TONG Qiang;YANG Da-li(College of Computer,Beijing Information Science and Technology University,Beijing 100101,China)

机构地区：[1]北京信息科技大学计算机学院,北京100101

出　　处：《液晶与显示》2022年第6期758-767,共10页Chinese Journal of Liquid Crystals and Displays

基　　金：北京市教委科研计划一般项目(No.KM202111232003);北京信息科技大学学科与研究生教育校内项目(No.5112111004)。

摘　　要：针对图像拼接的速度优化和最终匹配对优化问题,本文提出了一种基于BRISK(Binary Robust Invariant Scalable Keypoints)特征点检测算法和改进RANSAC(Random sample consensus)算法的图像拼接方法。改进的RANSAC算法旨在通过循环以求得更多匹配对,具体改进为在计算内点个数时加入循环以及将内点的判断时使用的欧氏距离改成面积。实验结果表明,本文改进的RANSAC算法使得内点个数增加,平均反向投影错误率相对于原算法减小了约10%;提出的拼接方法相对于常用的拼接方法时间缩短了约50%。本文提出的方法能够实现实时、准确的图像拼接。Aiming at the optimization of speed and of final matching pairs for image mosaicing,this paper proposes an image mosaicing algorithm based on BRISK(Binary Robust Invariant Scalable Keypoints)keypoint detection algorithm and improved RANSAC(Random sample consensus)algorithm.The improved RANSAC algorithm aims to find more matching pairs through loops.The specific improvements are adding loops while calculating the number of inliers and changing the Euclidean distance used for inliers judgment to area.The experimental results show that the improved RANSAC algorithm increases the number of inliers,and the average back-projection error rate is reduced by about 10%compared with the original algorithm.Compared with the common mosaicking methods,the time cost of the mosaicking method proposed in this paper is reduced by about 50%.The method proposed in this paper can realize real-time and accurate image mosaic.

关 键 词：图像拼接 改进RANSAC算法 单应性矩阵 BRISK算法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]