Clustered Redundant Keypoint Elimination SURF method in MRI Image Registration based on Alpha-Trimmed Relationship

Abstract

The process of MRI image registration is one of the important branches in MRI image analysis, which is a necessary preprocessing to the use of information in these images. The Clustered Adaptive Keypoint Elimination method-SIFT (CRKEM-SIFT) algorithm has recently been introduced to eliminate redundancies and upgrade the precision corresponding. The disadvantages of this algorithm are the high execution time and the number of incorrect correspondences. In this paper, to increase the accuracy and speed of MRI image registration, first, the CRKEM method is used on the SURF algorithm. Then, Spatial Relations Correspondence (SRC) and Alpha-Trimmed Spatial Relations Correspondence (ATSRC) methods are suggested to improve correspondences. These suggested methods, unlike conventional methods such as RANSAC, which only eliminates incorrect correspondences, in these suggested methods based on spatial relationships, detect incorrect correspondences and turn them into correct correspondences. Converting incorrect correspondences to correct ones can increase the number of correct correspondences and ultimately increase the precision correspondences. The simulation results confirm the suggested approaches' superiority on standard brain databases compared to classic methods in terms of Maximum error (MAE) and precision.

Keywords

• brain image registration
• CRKEM-SIFT
• spatial relations
• redundant keypoints
• MRI