DocumentCode :
3492868
Title :
A New Color Pattern Recognition Algorithm Based On Quaternion Correlation
Author :
Qi, Qifeng ; Hu, Yueming ; Zhang, Chaoyang
fYear :
2008
fDate :
6-8 April 2008
Firstpage :
462
Lastpage :
466
Abstract :
Quaternion correlation, which uses a quaternion to represent the R, G, B value of a pixel in color images, takes advantage of high recognizing precision and strong noise tolerance. However, it can hardly satisfy the speed requirement because of its convolution character. By the Fourier expression of quaternion correlation and the convolution theorem, the color pattern recognition problem is addressed in this paper, to enhance the implementation efficiency. The paper firstly introduces the basic knowledge about quaternion and quaternion correlation. Then, the detailed algorithm is proposed. Finally, we demonstrate the practical implication using example applications for color images, and prove the validity of the algorithm.
Keywords :
Fourier analysis; convolution; image colour analysis; image recognition; Fourier expression; color images; color pattern recognition algorithm; convolution character; convolution theorem; noise tolerance; quaternion correlation; Color; Colored noise; Convolution; Fourier transforms; Image processing; Image recognition; Manufacturing; Pattern recognition; Pixel; Quaternions;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Networking, Sensing and Control, 2008. ICNSC 2008. IEEE International Conference on
Conference_Location :
Sanya
Print_ISBN :
978-1-4244-1685-1
Electronic_ISBN :
978-1-4244-1686-8
Type :
conf
DOI :
10.1109/ICNSC.2008.4525261
Filename :
4525261
Link To Document :
بازگشت