基于二维粒子群优化的图像模糊增强算法研究

Research on Image Enhancement Algorithm Based on Fuzzy Enhancement Optimized by 2D Particle Swarm Optimization

针对传统基于灰度变换方法进行图像增强后图像质量不高等现象,对粒子群优化算法、模糊增强算法进行研究,同时结合禁忌搜素和粒子空间对称分布原理,提出一种基于二维粒子群优化的图像模糊增强算法。该算法通过对搜索粒子进行空间对称分布调整以避免算法陷入局部最优、提高全局搜索能力,并且在算法迭代后期加入禁忌搜索算法记录粒子搜索位置,以减少粒子位置重复寻优、提高算法搜索效率。最后将改进后的粒子群优化算法中粒子搜索位置和速度更新方向设定为二维并与模糊增强算法相结合,自适应搜索出模糊参数Fp和Fe最优值,实现模糊增强。实验结果表明,改进后算法对图像增强效果较好,并且将算法用于过暗SAR图像、医学MR图像的增强,可有效提高图像质量。

Aiming at the phenomenon of classic image enhancement method based on grayscale transformation method which is vulnerable to image quality, researching for the particle swarm optimization and fuzzy enhancement algorithm that an image enhancement algorithm based on fuzzy enhancement optimized by 2D particle swarm optimization is proposed which combined with tabu search and particle spatial distribution principle. In this algorithm, adjusting the particle distribution for spatial distribution of symmetry to avoid the algorithm falling into local optimal and improve the global search ability, and adding tabu search algorithm in traditional particle swarm optimization in order to reduce the duplication optimization of particles and improve the searching efficiency of the algorithm by recording the position of the particle search in the later of iteration. Finally, the particles＇ search position and velocity update direction of the improved particle swarm optimization algorithm are setted to the 2D and combined with fuzzy enhancement algorithm that can adaptive search the optimal value of fuzzy parameters Fp and Fc to achieve fuzzy enhancement. The experiments show that the improved algorithm is better for image enhancement and can effectively improve the image quality when used for SAR image and medical MR image enhancement that are too dar.