Sparse Representation of Image with Immune Clone Algorithm based on Harmonic Wavelet Packet Dictionary

Sparse representation of image could reduce the amount of original data and facilitate the subsequent processing. When using the redundant dictionary to represent the images, we need to look for an efficient optimal algorithm to seek the optimal atoms. In this paper, an immune clone algorithm for sparse representation of image based on harmonic wavelet packet dictionary is proposed. The proposed algorithm relies on the global optimal search ability of immune clone algorithm to realize the process of looking for optimal atoms. The fitness function for immune clone algorithm is designed and the implementation process for sparse representation is shown. Experimental results show that compared with orthogonal matching pursuit algorithm, the proposed algorithm could improve the computational efficiency while guarantee the reconstruction accuracy.