SIFT特征点引导的区域立体匹配算法  

Area-based stereo matching based on SIFT feature guidance

作　　者：郭龙源[1,2] 孙长银[1] 杨万扣[1] 张国云[2] 

机构地区：[1]东南大学自动化学院,南京210096 [2]湖南理工学院信息与通信工程学院,湖南岳阳414006

出　　处：《计算机工程与应用》2013年第4期23-25,38,共4页Computer Engineering and Applications

基　　金：国家自然科学基金青年项目(No.61005008);湖南省教育厅重点项目(No.10A046)

摘　　要：根据SIFT特征对旋转、尺度缩放、亮度变化保持不变性的特点,提出基于SIFT特征和边缘特征点的区域匹配方法。该方法确定符合SIFT特征的边缘为可靠特征点,并确定其视差;根据视差梯度原理确定其他点的视差,最后生成稠密的视差图。实验结果表明,SIFT特征的引入,提高了特征点视差的准确性,一些弱纹理区的匹配也有所改善。SIFT features is invariance in zoom, rotate, scale, brightness variations. According to these advantages, this paper presents a novel dense disparity matching algorithm based on the reliable feature points of SIFT and edge region. The method determines SIFT features points in the edge and their disparity; The other points＇ disparity are determined according to the principle of disparity gradient; A dense disparity map is generated. The experimental results show that the SIFT feature is to improve the accuracy of the disparity of feature points. And the disparity in some weak texture area has also been improved.

关 键 词：区域立体匹配 尺度不变特征变换 稠密视差图 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]