Title :
Research of Blind Images Separation Algorithm Based on Kernel Space
Author :
Chen, Lei ; Zhang, Liyi ; Guo, Yanju ; Liu, Ting
Author_Institution :
Sch. of Electron. Inf. Eng., Tianjin Univ., Tianjin, China
Abstract :
Principle of blind source separation (BSS) and kernel function method is introduced. Kernel method is a kind of new learning algorithm concerned by many scholars. More excellent new algorithm can be got by kernelizing the original algorithm using kernel trick. Kernelized blind source separation algorithm based on second-order statistics are expatiated and a new blind images separation algorithm using the kernel trick originally applied in support vector machine (SVM) is proposed. The result of experiment on realistic natural images shows that the blind images separation algorithm based on kernel space can separate mixed natural images successfully.
Keywords :
blind source separation; image processing; learning (artificial intelligence); support vector machines; blind images separation algorithm; kernel function method; kernel trick; kernelized blind source separation algorithm; learning algorithm; second-order statistics; support vector machine; Blind source separation; Independent component analysis; Kernel; Machine learning algorithms; Neural networks; Signal processing; Signal processing algorithms; Source separation; Space technology; Support vector machines; blind source separation; independent component analysis; kernel trick; support vector machine;
Conference_Titel :
Natural Computation, 2009. ICNC '09. Fifth International Conference on
Conference_Location :
Tianjin
Print_ISBN :
978-0-7695-3736-8
DOI :
10.1109/ICNC.2009.279