三维运动分析中的运动-立体双匹配约束  

Motion-stereo double matching restriction in 3D movement analysis

作　　者：张春森[1] 

机构地区：[1]西安科技大学测量工程系,陕西西安710054

出　　处：《光学精密工程》2007年第6期945-950,共6页Optics and Precision Engineering

基　　金：国家自然科学基金资助项目(No.40171080);陕西省教育厅专项科研计划项目(No.06KJ239)

摘　　要：为保证在基于点特征的三维运动分析中运动物体不同时刻任意物方特征点的对应,对该过程中存在的图像运动-立体匹配问题进行了研究。提出了运动-立体匹配相结合,分阶段的运动-立体双匹配约束的方法。即采用松弛法点特征匹配算法完成图像间的基本匹配,通过运动-立体双匹配约束完成运动物体特征点的对应。给出了一组真实的实验数据,结果表明:该算法总的正确率达76.5%,基本达到了三维运动物体定位跟踪中特征点对应的要求。In order to realize the correspondence of arbitrary object side feature points of moving object in the different time in three-dimensional （3D） movement analysis, motion-stereo image matching in the process is studied. A algorithm of double matching restriction and motion-stereo image of matching is presented. The basic image matching based on point feature is completed by correlation coeffi cients and relaxation algorithm,and the feature point correspondence of movwment object is performed by motion-stereo double matching restriction of binocular sequence image. A set of experiment results are presented,it shows that the accuracy of the final correspondence is about 76.5 %, which can meet the requirements of the 3D motion object tracking and locating the feature points.

关 键 词：图像匹配 序列图像 三维运动分析 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]