A robust feature-based registration method of multimodal image using phase congruency and coherent point drift

• DOI: 10.1117/12.2031615
• Publication Date: 2013-10-26T21:39:49Z
• Authors: Renbo Xia; Jibin Zhao; Yunpeng Liu
• Keywords: Scale-invariant feature transform; Phase congruency; Point set registration; Image registration; Feature (linguistics); Feature vector

This paper presents a new feature matching algorithm for nonrigid multimodal image registration. The proposed first constructs phase congruency representations (PCR) of images to be registered. Then scale invariant transform (SIFT) method is applied capture significant points from PCR. Subsequently, the putative obtained by nearest neighbour in SIFT descriptor space. then integrated into Coherent Point Drift (CPD) so that appropriate two point sets solved combining appearance with distance properties between match candidates. Finally, transformation estimated registration original images. results show increases correct rate and well suited multi-modal