基于小波变异粒子群和模糊熵的图像分割

Image segmentation based on wavelet mutation particle swarm optimization and fuzzy entropy

基于粒子群和模糊熵的图像分割算法用于各种图像分割时,由于基本粒子群算法存在易陷入局部最优以及过早收敛的缺点,使得该算法难以得到理想的分割效果。针对此问题,提出了一种基于小波变异粒子群和模糊熵的图像分割算法,利用小波变异粒子群来搜索使模糊熵最大的参数值,得到模糊参数的最优组合,进而确定图像的分割阈值。通过与其他两种粒子群算法的分割结果进行比较,表明该算法取得了令人满意的分割结果,算法运算时间较小,具有很好的自适应性。

Image segmentation algorithm based on Particle Swarm Optimization （PSO） and fuzzy entropy cannot have satisfactory performance because the classic PSO easily falls into local optimization and premature convergence. Concerning this shortcoming, a new segmentation algorithm based on wavelet mutation PSO and fuzzy entropy was proposed. The new algorithm used wavelet mutation PSO to explore fuzzy parameters of maximum fuzzy entropy, and to get the optimum fuzzy parameter combination, and then obtained the segmentation threshold. According to the comparison of the experimental results between the new algorithm and the other two algorithms, the new algorithm is of good segmentation, low time cost and self adaptivity.