量子行为粒子群算法在图像恢复中的应用

摘要

遗传算法等智能搜索技术避免了图像恢复方法中存在的较多约束和计算量过大的问题，但遗传算法存在“过早收敛”现象。作为一种新的智能优化算法-量子行为粒子群优化算法，在全局收敛性和稳定性上有较好的表现。文章提出了一种基于量子行为粒子群算法的图像恢复方法，并与基于标准遗传算法的图像恢复进行了比较。仿真结果表明，该算法可使图像恢复结果和效率得以较大的改善和提高，具有推广应用价值。%The intelligent search technologies, such as genetic algorithms, avoid the problems of too many restrictions and complex computation in image restoration, but genetic algorithms may early convergent. Quantum-behaved Particle Swarm Optimization as a new method of optimization is better in convergence and stability of the overall. The article presents a method based on quantum-behaved particle swarm optimization in image restoration, compared with the method of genetic algorithms in image restoration. The simulation results indicate that this scheme can improve the performance and efifciency of image restoration, and that the algorithm is of great value in extension and application.