基于Lévy飞行的改进飞蛾扑火算法优化红外图像分割

Improved Moth-Flame Optimization Algorithm Based on Lévy Flight to Optimize Infrared Image Segmentation

针对使用传统阈值分割方法对电力设备故障诊断效率低、精度低的问题,使用智能算法优化Otsu算法对红外图像进行阈值分割再进行故障诊断。根据基本飞蛾扑火(Moth-Flame Optimization,MFO)算法缺点提出改进飞蛾扑火算法(Improved Moth-Flame Optimization Algorithm,IMFO)并将其应用红外图像分割中,通过对比粒子群算法(Particle Swarm Optimization,PSO)、生物地理算法(Biogeography-Based Optimization,BBO)、基本飞蛾扑火算法红外图像分割效果,表明改进算法取得成功。提出一种通过温度区域对红外图像进行多阈值分割的方法,能够准确确定每个部分的温度范围,从而保证设备的正常运行。

To solve the problem of low efficiency and accuracy of power equipment fault diagnosis using the traditional threshold segmentation method,an intelligent algorithm,the optimized Otsu algorithm was used for threshold segmentation of infrared images for fault diagnosis.According to the shortcomings of the basic moth-flame optimization,the improved moth-flame optimization algorithm is proposed.It was applied to the infrared image segmentation.By comparing its infrared image segmentation results with those of the particle swarm optimization,biogeography-based optimization,and moth–flame optimization algorithms,it was shown that the improved algorithm is successful.A multithreshold segmentation method for infrared images through the temperature region is proposed.It can accurately determine the temperature range of each part and ensure normal operation of the equipment.