基于图割的图像匹配算法

用标号表示视差,建立能量函数,把匹配问题转化为能量函数最小化问题;通过构造网络,使能量与网络的割的容量相联系;利用图的网络流理论给出能量函数的最小化,从而获得图像匹配的视差数据.与目前已有基于图割的匹配算法相比,本算法将标号从1维向量推广到2维向量,适用于更一般情形下的视觉匹配,并且在全局上获得能量函数最小.实验结果表明,所提的匹配算法准确率较高.

An Images Matching Algorithm Based on Graph Cuts

Label is denoted by disparity and the energy fimction is established. Then the problem of matching can be transformed into that of energy function minimization. A network is constructed such that the energies can be related to the capacities of the cuts of the network. Finally, the minimal energy is obtained by the network-flows theory, and hence the disparity data are solved. Comparing with some known algorithms based on graph cuts, the algorithm in this paper extends the label from 1 dimension vector to 2 dimension vector, and adapts vision matching of more general conditions ;furthermore the algorithm can gain the minimization in global. Experimental results show that the algorithm has a high accuracy.