Image Matching Based on Improved Harris Criterion Algorithm

Journal of Multimedia

关键词: Graph Transformation Matching;    Harris Criterion;    SIFT Features;    Feature Points Extraction;    Image Matching;   

Others  :  1017301
DOI  :  10.4304/jmm.9.1.145-151

【 摘 要 】

This paper proposes a local invariant feature for image matching based on Harris threshold criterion. Based on the extraction of SIFT invariant features, this algorithm chooses the effective features from the extracted invariant features utilizing Harris threshold criterion, which removes lots of feature points with poor distinction, thus gains relatively stable and better distinctive feature points. Secondly, complete the precise matching between the feature point sets combined the invariant feature vector with graph transformation matching method. Experimental results show the feasibility and robustness of this method in image matching

【 授权许可】

   
@ 2006-2014 by ACADEMY PUBLISHER – All rights reserved.