基于欧式距离约束的异源图像融合算法研究  

Research on Heterologous Image Fusion Algorithm Based on Euclidean Distance Constraint

作　　者：刘达 杜世逾 侯忠 黄翔 王启东 LIU Da;DU Shiyu;HOU Zhong;HUANG Xiang;WANG Qidong

机构地区：[1]国家能源集团新疆吉林台水电开发有限公司,新疆伊犁835700

出　　处：《电力系统装备》2025年第2期182-184,共3页Electric Power System Equipment

摘　　要：针对光伏组件红外光图像与可见光图像在同一坐标系下无法对齐,而传统异源图像融合算法在融合过程中存在图像细节特征损失过多、易出现特征点误匹配的问题,文章提出一种基于欧式距离约束的光伏组件红外图像与可见光图像融合算法。该算法利用Harris角点检测提取图像特征点,在特征点匹配过程中引入欧式距离准则进行约束,以获取更多的有效匹配特征点,从而提高匹配准确度,实现双光图像的融合。试验结果表明,文章提出的基于欧式距离约束的光伏组件双光图像融合算法不仅能够有效提高图像融合效果,在配准效率上也有所提升,满足大型光伏电站的实际需求。In view of the fact that the infrared image of photovoltaic modules and the visible image cannot be aligned in the same coordinate system,and the traditional heterologous image fusion algorithm has the problem of excessive loss of image details and mismatching of feature points in the fusion process,this paper proposes a fusion algorithm of infrared image and visible light image of photovoltaic module based on Euclidean distance constraint.Firstly,Harris corner detection is used to extract image feature points,and then the Euclidean distance criterion is introduced into the feature point matching process to obtain more effective matching feature points,so as to improve the matching accuracy and realize the fusion of dual-light images.Experimental results show that the dual-optical image fusion algorithm of photovoltaic modules based on Euclidean distance constraint proposed in this paper can effectively improve the image fusion effect,and also improve the registration efficiency,so as to meet the actual needs of large-scale photovoltaic power stations.

关 键 词：图像融合 欧式距离 特征点匹配 角点检测 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]