基于小波变换的PSO优化紫外双通道图像融合研究  

Research on PSO Optimization of Ultraviolet Dual-Channel Image Fusion Based on Wavelet Transform

作　　者：韩勇勇 HAN Yongyong(School of Electronics and Information,Xi’an Polytechnic University,Xi’an 710048,China)

机构地区：[1]西安工程大学电子信息学院,陕西西安710048

出　　处：《中国照明电器》2025年第2期95-102,共8页China Light & Lighting

摘　　要：将紫外(UV)光作为高压设备放电检测对象,能够更及时、准确地获取设备的运行状态;以可见光为背景,叠加紫外放电图像的融合结果相比单一的紫外图像更直观、准确。由于紫外与可见光属于异类图像,很难获取相似的特征点进行融合。本文提出使用粒子群优化算法(PSO)优化小波图像融合的方法,对于包含图像边缘信息的低频分量,使用加权融合算法;对于细节众多的高频分量使用局部方差的算法进行融合,利用PSO算法优化方差的求解过程,获取双通道的最佳融合权重,通过对不同距离下融合图像的参数提取,验证了本文算法的可行性,融合精度可达97%。Using ultraviolet(UV)light as the discharge detection object of high-voltage equipment can obtain the operation status of the equipment in a more promptly and accurate manner.With visible light as the background,the fusion results of superimposed UV discharge images are more intuitive and accurate than single UV images.Since ultraviolet and visible light are heterogeneous images,it is difficult to obtain similar feature points for fusion.In this paper,we propose a method to optimize Wavelet image fusion using Particle swarm optimization(PSO)algorithm,and use a weighted fusion algorithm for the low-frequency component components containing image edge information.For the high-frequency components with many details,the local variance algorithm is used for fusion,and the PSO algorithm is used to optimize the solution process of variance,and the optimal fusion weight of the two channels is obtained,and the feasibility of the proposed algorithm is verified by extracting the parameters of the fused images at different distances,and the fusion accuracy can reach 97%.

关 键 词：紫外光 图像融合 PSO 小波图像融合 

分 类 号：TN23[电子电信—物理电子学] TM835.1[电气工程—高电压与绝缘技术]