Singular value feature extraction of color image and its application for recognition

A new tool of feature extraction and recognition of color image is introduced. It is based on the quaternion representation of color image and quaternion matrix algebra, which are discussed firstly in the paper. Then the singular value (SV) feature vector of a color image is introduced and its many important properties are proved. The properties are very useful for the description and recognition of color image. Then a method for color image recognition is presented. As an application of SV feature vector, the experiment of color image recognition is made and the experiment results are satisfactory.

[1]  Zi-Quan Hong,et al.  Algebraic feature extraction of image for recognition , 1991, Pattern Recognit..

[2]  Soo-Chang Pei,et al.  Quaternion matrix singular value decomposition and its applications for color image processing , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[3]  Stephen J. Sangwine,et al.  Color image decomposition using quaternion singular value decomposition , 2003 .

[4]  Wang Yun.  Face Identification Based on Singular Value Decomposition and Data Fusion , 2000 .

[5]  Soo-Chang Pei,et al.  A novel block truncation coding of color images by using quaternion-moment-preserving principle , 1996, 1996 IEEE International Symposium on Circuits and Systems. Circuits and Systems Connecting the World. ISCAS 96.

[6]  Fuzhen Zhang.  Quaternions and matrices of quaternions , 1997 .

[7]  J Shuai,et al.  Hamilton neural-network model: recognition of the color patterns. , 1995, Applied optics.

[8]  Soo-Chang Pei,et al.  Color pattern recognition by quaternion correlation , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).