KFCE: A dictionary generation algorithm for sparse representation

Sparse representation (SR) for signals over an overcomplete dictionary fascinates a lot of researchers in the past decade. Using an overcomplete dictionary that contains prototype signal—atoms, signals are described by sparse linear combinations of these atoms. This paper addresses the problem of dictionary generation in SR. Recent studies show that this problem is equivalent to the problem of codebook estimation in vector quantization (VQ). A kernel fuzzy codebook estimation (KFCE) algorithm is proposed in this paper. The principle of the KFCE algorithm is to integrate the distance kernel trick with the fuzzy clustering algorithm to generate dictionary for SR. Experimental results on real image data show that the KFCE is fit for generating dictionary for SR.