海洋测绘Issue(4)：48-54,7.DOI:10.3969/j.issn.1671-3044.2025.04.010
基于多尺度引导共现滤波的红外-可见光图像融合
Infrared and visible image fusion based on multiscale guided and co-occurrence filtering
摘要
Abstract
Aiming at the problem that it is difficult to simultaneously reflect the strong edges of infrared images and the texture details of visible light images in infrared and visible images fusion,which leads to the unremarkable targets in the fused images and the unclear edges and textures.This paper proposes a hybrid multiscale decomposition and hierarchical fusion strategy achieved by combining guided and co-occurrence filtering.Through the hybrid multiscale decomposition,the source images are decomposed into a base layer,a texture detail layer,and an edge detail layer,realizing the refined separation of spatial superposition features.A fusion strategy is designed specifically according to the distribution of image features in different scale layers.The fused base layer,small scale layer,and large scale layer are linearly superimposed to obtain the final fused image.By conducting experiments in five typical scenes and comparing and analysing with other six fusion algorithms,the algorithm proposed in this paper has the most outstanding performance in subjective vision,with the best edge extraction and texture detail retention,and six objective evaluation indexes are in the leading edge,which fully verifies the effectiveness of this paper's algorithm.关键词
图像融合/多尺度分解/多层级融合/引导滤波/共现滤波Key words
image Fusion/multiscale decomposition/multilevel fusion/guided filtering/co-occurrence filtering分类
天文与地球科学引用本文复制引用
王颖颖,王怡衡,赵育良,郭鸿基,张凤明..基于多尺度引导共现滤波的红外-可见光图像融合[J].海洋测绘,2025,(4):48-54,7.基金项目
国家自然科学基金资助项目(91538201). （91538201）
