Title :
Selection Method for Scene Matching Area Based on Information Entropy
Author :
Xiaochen Zhang ; Zhongwen He ; Yihui Liang ; Peng Zeng
Author_Institution :
Syst. Eng. Res. Inst., China State Shipbuilding Corp., Beijing, China
Abstract :
A new selection method for the Scene Matching Area(SMA) based on information entropy is proposed. to improve the reliability in selecting the SMA, there scene matching ability parameters included image entropy, Frieden entropy and normalized mutual information are defined based on entropy theory to evaluate the information content, stability and unique characteristic of the SMA. According to that, the best SMA is selected step by step. Experiment results validate that the SMA selected by this method has high matching probability and low matching errors. the method, which is logical and reasonable, can meet the matching requirements.
Keywords :
entropy; image matching; natural scenes; probability; Frieden entropy; SMA characteristic evaluation; SMA information content evaluation; SMA selection reliability improvement; SMA stability evaluation; image entropy; information entropy; matching errors; matching probability; normalized mutual information; scene matching ability parameters; scene matching area selection method; Correlation; Entropy; Indexes; Information entropy; Mutual information; Stability criteria; information entropy; mutual information; scene matching; selection rule of SMA;
Conference_Titel :
Computational Intelligence and Design (ISCID), 2012 Fifth International Symposium on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4673-2646-9
DOI :
10.1109/ISCID.2012.98
