科研之友
微信
新浪微博
Facebook
分享链接
摘要
Various feature point matching approaches have been reported in computer vision. However, few feature point matching approaches employed on images from non-rigid and non-uniform biological tissues have been reported. The present study investigates the interior deformation field measurement of complex biological tissues from three-dimensional magnetic resonance (MR) volumetric images. To improve the reliability of the matching, we propose the fusion of three different similarity measurement methods. The basic algorithm of the three different similarity measurement methods and their fusion to generate an integration method to improve the results are discussed. The validity of the proposed approach was tested by applying it to actual MR volumetric images that captured the calf of a volunteer. The results indicate that the fusion method is more effective than the single feature point matching approach.
