An efficient volumetric matching algorithm based on MSVs and 3DSURF

Volumetric matching is an important tool in volume image analysis. In this paper, we propose an efficient volumetric matching framework to extract coarse correspondences between given volume images. The framework includes two steps. First, we extract features based on a Maximally Stable Volumes (MSVs) algorithm then build feature descriptors using a scale-related 3DSURF algorithm reliable to scale changes. Then, we match features using a Spectral Matching (SM) algorithm. We show the effectiveness and robustness of our algorithm on both synthetic and clinical medical images.