基于混沌电磁场优化算法的多阈值彩色图像分割

MULTI-THRESHOLD COLOR IMAGE SEGMENTATION BASED ON CHAOTIC ELECTROMAGNETIC FIELD OPTIMIZATION

为了解决彩色图像多阈值分割中计算时间长、分割精度低的问题,在电磁场优化算法(Electromagnetic Field Optimization,EFO)的基础上引入一种混沌策略用于算法初始化中,提出混沌电磁场优化算法(Chaotic Electromagnetic Field Optimization,CEFO)对图像的最佳阈值向量进行搜索。将其与另外5种优化算法进行对比,采用PSNR、MSSIM和FSIM 3个图像质量评价指标和算法运行时间(CPU Time)对6种分割算法进行分析比较。结果表明,CEFO具有收敛速度快、分割精度高的优势,能够胜任多阈值彩色图像分割的工程任务。

In order to solve the problem of long computation time and low segmentation accuracy in multi-threshold segmentation of color images,this paper introduces a chaotic strategy based on electromagnetic field optimization(EFO)to initialize the algorithm,and proposes chaotic electromagnetic field optimization(CEFO)to search the optimal threshold vector of the image.It was compared with the other five optimization algorithms.PSNR,MSSIM and FSIM were used to evaluate the image quality,and the computational time(CPU Time)of six segmentation algorithms was analyzed and compared.The results show that CEFO has the advantages of fast convergence and high segmentation accuracy,and can be competent for the engineering task of multi-threshold color image segmentation.