【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。更多还原

【Abstract】 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.更多还原