Adaptive Annealing Genetic Algorithm for Wavelet Denoising

It´s very difficult to select the best wavelet denoising threshold. A novel adaptive annealing genetic algorithm is presented to improve convergence and stability of standard genetic algorithm. A new adaptive annealing method is given to calculate select probability for improving the convergence of this algorithm. Cross probability and variance probability are selected adaptively for enhancing this algorithm stability and convergence. The convergence of this algorithm can be ensured by competition in male parent1. There are many merits such as convergence rapidly, avoiding local extremum and global optimization ability in this algorithm. The simulation shows that the best wavelet denoising threshold parameter can be found effectively by this algorithm.