基于RANSAC算法的基本矩阵估计的匹配方法  

The Matching Method Based on RANSAC Algorithm for Estimation of the Fundamental Matrix

作　　者：单欣[1] 王耀明[2] 董建萍[1] 

机构地区：[1]上海师范大学数理信息学院,上海200234 [2]上海电机学院电子信息学院,上海200240

出　　处：《上海电机学院学报》2006年第4期66-69,共4页Journal of Shanghai Dianji University

基　　金：上海市教委科研项目(沪教科01D02-1)

摘　　要：基本矩阵包含了摄像机的所有内参数和外参数信息,求解基本矩阵是计算机视觉中的重要研究课题。在介绍极线几何和基本矩阵理论后,提出了一种基于Sampson距离的RANSAC(Ran-dom Sampling Consensus)算法用于解决求解基本矩阵中误匹配问题。通过从图像的特征提取、特征点匹配到求解基本矩阵的完整的仿真实验,验证了该算法的有效性。The fundamental matrix contains all of a camera＇s inner and outer parameters of a camera. So the estimation of the fundamental matrix is a crucial problem of computer vision. In this paper, we introduce the theorem of the epipolar geometry and the fundamental matrix. To solve the problem of false matches, we also introduces RANSAC algorithm based on Sampson distance. Experiments on feature detection and matching as well as the estimation of the fundamental matrix show that the proposed method is effective.

关 键 词：基本矩阵 极线几何 RANSAC方法 Sampson距离 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]