摘要
提出了一种基于特征点的单应矩阵鲁棒估计算法。在图像的尺度空间中提取特征点,并对特征点进行亚像素定位,同时赋予主方向。根据邻域信息计算得到特征向量后,利用最近邻特征点距离与次近邻特征点距离之比得到初始匹配点对。用RANSAC(Random Sample Consensus)算法匹配特征点对,同时计算得到两幅图像之间的单应矩阵。试验结果表明,该算法能够实现单应矩阵的鲁棒估计。
Proposing a homography robust estimation algorithm based on feature points.extracts the feature points from the scale space of the image,and locates feature points on the sub-pixel coordinates,and gives the main orientation.Initial feature points matching can be calculated by using the distance ratio of the nearest neighbor feature point and the next nearest neighbor feature point.Then uses the RANSAC(Random Sample Consensus) algorithm to match the initial feature points set,while calculate the homography between the two images.The results show that the method can achieve robust homography estimation.
出处
《电脑编程技巧与维护》
2011年第10期33-34,47,共3页
Computer Programming Skills & Maintenance
基金
西南林业大学面上基金项目(编号20616M)
关键词
特征提取
特征匹配
单应矩阵
鲁棒估计
feature extraction
feature matching
homography
robust estimation