DocumentCode :
2963258
Title :
A New Image Registration Method Based on Point Pattern
Author :
Zhi-gang, Chen ; Jun, Jiang
Author_Institution :
Electron. Eng. Coll., Naval Eng. Univ., Wuhan, China
Volume :
2
fYear :
2011
fDate :
28-29 March 2011
Firstpage :
354
Lastpage :
357
Abstract :
Data registration is the most important segment in the data fusion process. The image registration problem was firstly analyzed in this paper. Then it uses the interest point detection operator to extract the point pattern, use centroids of these features to define initial matching pairs. By constructing point pattern triangles and uses its similarity to define support degree and match degree, a new six-element point pattern matching algorithm based on initial matching pairs was proposed. Under the condition of scale and rotation, the algorithm achieved the matching correctly.
Keywords :
feature extraction; geometry; image matching; image registration; sensor fusion; data fusion process; data registration; image registration method; point pattern extraction; point pattern matching algorithm; point pattern triangles; Feature extraction; Hopfield neural networks; Image registration; Pattern matching; Wavelet domain; Wavelet transforms; centroid of feature; image registration; six-element point pattern matching algorithm;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Computation Technology and Automation (ICICTA), 2011 International Conference on
Conference_Location :
Shenzhen, Guangdong
Print_ISBN :
978-1-61284-289-9
Type :
conf
DOI :
10.1109/ICICTA.2011.372
Filename :
5750897
Link To Document :
بازگشت