Tuning an Iterated Local Search Algorithm for Wavelet Sign Coding for 2D Image Compression

Wavelet transforms have proved to be very powerful tools for image compression. Previous studies have verified that there is a strong correlation between the sign of a wavelet coefficient and the signs of their neighbors. This correlation opens the possibility of using a sign predictor in order to improve the image compression process. In this paper a new sign coding approximation method for the wavelet coefficients in a 2D image codec based on an iterated local search algorithm (ILS) is presented. The efficiency of the proposed algorithm versus simulated annealing algorithm (SA) and genetic algorithm (GA) using standard images and benchmarks of Kodak is compared. The proposed sign prediction algorithm is as efficient as other methods and it provides a significant reduction of wavelet coefficients sign information in the final bit-stream.