《2D registration based on contour matching for partial matching images》.pdfVIP

J. Cent. South Univ. (2014) 21: 4553−4562 DOI: 10.1007/s11771-014-2460-z 2D registration based on contour matching for partial matching images ZHANG Jian-wei(张见威), HUANG Da-cheng(黄达承), GUI Jiang-qin(桂姜琴), YE Wen-zhong(叶文忠) School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China © Central South University Press and Springer-Verlag Berlin Heidelberg 2014 Abstract: The mean Hausdorff distance, though highly applicable in image registration, does not work well on partial matching images. An improvement upon traditional Hausdorff-distance-based image registration method is proposed, which consists of the following two aspects. One is to estimate transformation parameters between two images from the distributions of geometric property differences instead of establishing explicit feature correspondences. This procedure is treated as the pre-registration. The other aspect is that mean Hausdorff distance computation is replaced with the analysis of the second difference of generalized Hausdorff distance so as to eliminate the redundant points. Experimental results show that our registration method outperforms the method based on mean Hausdorff distance. The registration errors are noticeably reduced in the partial matching images. Key words: image registration; generalized Hausdorff distance; partial matching image used in MRI image fusion by registering a proton density 1 Introduction image with its T1 weighted image. Owing to the convenience of their implementation, direct 2D-to-2D Image registration is the process of a