Medical Image Registration Combined with SURF and Improved RANSAC Algorithm

Zong Yun Gu; Chun Min Du; Li Jin; Hong Chun Tan

doi:10.4028/www.scientific.net/AMM.411-414.1233

Paper Titles

The Study and Implementation of Digital Camera Positioning Based on Pinhole Model
p.1213

Fault Automatic Detection Method of Steel Cord Conveyor Belt Based on Gabor Filter Bank
p.1218

Watershed Segmentation Based on Histogram Threshold and Lowest Mean Edge Value
p.1222

Research on Facial Expression Recognition Based on Variable Precision Fuzzy Rough Set
p.1228

Medical Image Registration Combined with SURF and Improved RANSAC Algorithm
p.1233

Printed Thai Character Recognition Using Conditional Random Fields and Hierarchical Centroid Distance
p.1238

Optimization of K-Means Clustering Segmentation in Head CT Images
p.1247

Spectrum Distribution of Landuse Classification by Genetic Algorithm
p.1251

Experimentally Determined the Structure of Crowd Formation Recognition Model with K2
p.1256

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 411-414Medical Image Registration Combined with SURF and...

Medical Image Registration Combined with SURF and Improved RANSAC Algorithm

Abstract:

Aiming at the requirements of medical image registration for good robustness, high-accuracy and speed, this paper proposes a Medical Image Registration algorithm Combined SURF with improved RANSAC algorithm. This algorithm first of all extracts SURF featured points from images and matches similar featured points, then the improved RANSAC algorithm is used to eliminate wrong matches, and finally the image registration process is accomplished by estimating space geometric varied parameters according to least square method. The algorithm combines robustness and high efficiency of SURF and high-accuracy of improved RANSAC algorithm. Experimental results show that in the condition of images with noise, non-uniform intensity and large scope of the initial misalignment, the proposed algorithm achieves better robustness and higher speed while maintaining good registration accuracy compared with the conventional ional area-based and feature-based registration methods.