基于红外可见光特征增强与融合的目标检测算法  

An Object Detection Algorithm Based on Infrared-Visible Feature Enhancement and Fusion

作　　者：李明禄 王肖霞[1,2] 侯茂新 杨风暴 FusionLI Minglu;WANG Xiaoxia;HOU Maoxin;YANG Fengbao(College of Information and Communications Engineering,North University of China,Taiyuan 030051,China;Key Laboratory of Intelligent Information Control Technology of Shanxi Province,Taiyuan 030051,China;Collective Intelligence&Collaboration Laboratory,Zhongbing Intelligent Innovation Research Institute Limited Liability Company,Beijing 100072,China)

机构地区：[1]中北大学信息与通信工程学院,山西太原030051 [2]智能信息控制技术山西省重点实验室,山西太原030051 [3]中兵智能创新研究院有限公司群体协同与自主实验室,北京100072

出　　处：《红外技术》2025年第3期385-394,共10页Infrared Technology

摘　　要：为了应对复杂动态环境下红外与可见光双模态目标检测的挑战,特别是目标特征表达不足以及红外可见光特征在双模态融合中无法充分利用互补特征导致漏检和误检的问题,提出了一种用于目标检测的双分支特征增强与融合网络(Dual-Branch Feature Enhancement and Fusion,DBEF-Net)。针对模型对红外和可见光特征关注度不足的问题,设计了一种特征交互增强模块,该模块能够有效地关注并增强双模态特征中的有用信息。同时,为了更有效地利用双模态的互补特征,采用基于Transformer的双模态融合网络,并引入交叉注意力机制,以实现模态间的深度融合。实验结果表明,在SYUGV数据集上,与现有双模态目标检测算法相比,本文方法的平均检测精度更高,处理速度也能满足实时检测的需求。A dual-branch feature enhancement and fusion backbone network(DBEF-Net)is proposed for object detection to address the challenges of infrared and visible bimodal object detection in complex dynamic environments.Specifically,DBEF-Net addresses issues such as insufficient object feature expression and the inability of infrared and visible features to fully utilize the complementary features in bimodal fusion leading to omission and misdetection.To further address the insufficient attention of the model to infrared and visible light features,a feature interaction enhancement module is designed to effectively focus on and enhance the useful information in bimodal features.A transformer-based bimodal fusion network is further adopted.To utilize the complementary features of bimodal modalities more effectively,a cross-attention mechanism is introduced to achieve deep fusion between the modalities.Experimental results show that the proposed method has higher average detection accuracy than existing bimodal object detection algorithms on the SYUGV dataset,meeting the processing speed for real-time detection.

关 键 词：双模态图像 目标检测 特征融合 深度学习 注意力机制 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]