DocumentCode
442204
Title
A new point matching method based on position similarity
Author
Pan, Jun-Jun ; Zhang, Yan-ning
Author_Institution
Sch. of Comput. Sci., Northwestern Polytech. Univ., Xi´´an, China
Volume
8
fYear
2005
fDate
18-21 Aug. 2005
Firstpage
5154
Abstract
A new corresponding point matching method is presented when studying the feature points matching from X-ray images, which is called "the regulation of the minimum summation of Euclid distance". This method is different from the conventional matching approaches based on gray level or based on region geometric feature. It is based on the position similarity of corresponding points. This method derives from the model of sequence matching algorithm. According to the condition that the relative position of any two points in adjacent area from two correlative images are almost constant, this method minimizes the summation of corresponding points distance by adjusting the sequence of points through evolutionary programming searching. The experimental result shows that this method can match the most feature points correctly in low time-consuming, just based on the position similarity of points.
Keywords
X-ray imaging; evolutionary computation; feature extraction; image matching; image sequences; minimisation; search problems; Euclid distance summation; X-ray image; evolutionary programming searching; feature points matching; minimization; point matching; position similarity; sequence matching algorithm; Corresponding points matching; Euclid Distance; Evolutionary Programming; Position similarity;
fLanguage
English
Publisher
ieee
Conference_Titel
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location
Guangzhou, China
Print_ISBN
0-7803-9091-1
Type
conf
DOI
10.1109/ICMLC.2005.1527852
Filename
1527852
Link To Document