Improved QPSO Algorithm And Its Application

The purpose of this paper is to research Particle Swarm Optimization algorithm (PSO for short), especially Quantum-behaved Particle Swarm Optimization algorithm (QPSO for short). And improve QPSO algorithm based on the research, then apply the improved QPSO algorithm to image enhancement. In image enhancement, non-linear transform of gray level image is a very valid method. The essence of this method is making some transform of image sequence in order to achieve the purpose of image enhancement. It is a problem to confirm the parameters in the processes of transform, so it is very meaningful to establish a self-adapting algorithm for image enhancement which bases on the auto-adapting character of gray level image.Firstly, it demonstrates the main theory of image enhancement and the thinking of PSO algorithm and QPSO algorithm in the paper. Then, we make a comparison between these two algorithms; emphasize the better performance of QPSO algorithm in global convergence.Secondly, we put forward an improved QPSO algorithm by introducing a new searching strategy based on QPSO algorithm. Results of experiments show the improved QPSO algorithm performs better in global convergence than PSO algorithm and QPSO algorithm, and the larger the swarm is, the better the performance is.Finally, for non-complete Beta function which can cover all the typical transform types in image enhancement, we use this improved QPSO algorithm to achieve the self-adapting choice of its parameters. Many results of experiments prove that for gray level image enhancement, improved QPSO algorithm performs better than PSO algorithm and QPSO algorithm.