中国光学（中英文）2025，Vol.18Issue(2)：317-332,16.DOI:10.37188/CO.2024-0147
基于跨域交互注意力和对比学习引导的红外与可见光图像融合
Infrared and visible image fusion guided by cross-domain interactive attention and contrastive learning
摘要
Abstract
Aiming at the problems in existing infrared and visible image fusion methods,such as the diffi-culty in fully extracting and preserving the source image details,contrast,and blurred texture details,this pa-per proposes an infrared and visible image fusion method guided by cross-domain interactive attention and contrastive learning.First,a dual-branch skip connection detail enhancement network was designed to separ-ately extract and enhance detail information from infrared and visible images,using skip connections to pre-vent information loss and generate enhanced detail images.Next,a fusion network combining a dual-branch encoder and cross-domain interactive attention module was constructed to ensure sufficient feature interac-tion during fusion,and the decoder was used to reconstruct the final fused image.Then,a contrastive learn-ing network was introduced,performing shallow and deep attribute and content contrastive learning from the contrastive learning block,optimizing feature representation,and further improving the performance of the fusion network.Finally,to constrain network training and retain the inherent features of the source images,a contrast-based loss function was designed to assist in preserving source image information during fusion.The proposed method is qualitatively and quantitatively compared with current state-of-the-art fusion methods.Experimental results show that the eight objective evaluation metrics of the proposed method significantly outperform the comparison methods on the TNO,MSRS,and RoadSence datasets.The fused images pro-duced by the proposed method have rich detail textures,enhanced sharpness,and contrast,effectively im-proving target recognition and environmental perception in real-world applications such as road traffic and security surveillance.关键词
红外与可见光图像融合/对比学习/跨域交互注意力机制/对比约束损失Key words
infrared and visible image fusion/contrastive learning/cross-domain interactive attention mech-anisms/contrast constraint loss分类
信息技术与安全科学引用本文复制引用
邸敬,梁婵,刘冀钊,廉敬..基于跨域交互注意力和对比学习引导的红外与可见光图像融合[J].中国光学（中英文）,2025,18(2):317-332,16.基金项目
甘肃省自然科学基金项目(No.24JRRA231) （No.24JRRA231）
国家自然科学基金(No.62061023) （No.62061023）
甘肃省杰出青年基金资助项目(No.21JR7RA345) Supported by Natural Science Foundation of Gansu Province(No.24JRRA231) （No.21JR7RA345）
National Natural Science Foundation of China(No.62061023) （No.62061023）
Gansu Provincial Outstanding Youth Fund Project(No.21JR7RA345) （No.21JR7RA345）
