光学精密工程2025，Vol.33Issue(8)：1303-1312,10.DOI:10.37188/OPE.20253308.1303
感知增强混合网络的水下目标检测
Perception enhanced hybrid network for underwater object detection
摘要
Abstract
Underwater object detection technology plays an important role in areas of marine resource ex-ploration and environmental protection.However,the problems of blurred imaging and variable object scales in underwater environments pose difficulties for detection tasks.As a result,it is challenging for ac-curate underwater object feature extraction,which influences the detection performance of existing meth-ods.To solve the above-mentioned problem,a feature enhanced hybrid network was proposed to improve the detection accuracy of underwater objects.Firstly,a global-local hybrid feature enhancement network was constructed.The long-range global information in the image was extracted via self-attention mecha-nisms,and the richer local detailed information was further calculated through the devised convolutional at-tention enhancement module.The global and local relationships in the images could be better established,hence the multiscale feature representation powers of the network were enhanced.Subsequently,in order to suppress the interference of imaging blurriness and low contrast in underwater environments,a two-stage object perception enhanced detection head was constructed.The depth of the first-stage region pro-posal network was increased,thus more semantic information of underwater objects could be extracted.Besides,the self-attention mechanism was introduced in the second stage to suppress the interference from background noise.Moreover,an intersection-over-union branch was added to further integrate the prior in-formation of objects obtained from the first stage into the second stage.The proposed method achieves 37.8%,61.8%,and 82.0%,98.9%of mAP0.5:0.95 and AP50 on the TrashCan and WPBB datasets re-spectively.The qualitative and quantitative comparison experimental results demonstrate that this method could obtain superior detection results for various underwater objects.关键词
水下目标检测/特征增强/自注意力机制/混合网络Key words
underwater object detection/feature enhancement/self-attention/hybrid network分类
计算机与自动化引用本文复制引用
姚婷婷,李宁,张煜..感知增强混合网络的水下目标检测[J].光学精密工程,2025,33(8):1303-1312,10.基金项目
国家自然科学基金(No.62001078) （No.62001078）
中央高校基本科研业务费(No.3132025245) （No.3132025245）
