基于多尺度分层残差网络的光学遥感图像微小目标检测  

A Multi-scale Hierarchical Residual Network-based Method for Tiny Object Detection in Optical Remote Sensing Images

作　　者：曾祥津 刘耿焕 陈建明 豆嘉真 任振波 邸江磊 秦玉文 ZENG Xiangjin;LIU Genghuan;CHEN Jianming;DOU Jiazhen;REN Zhenbo;DI Jianglei;QIN Yuwen(Key Laboratory of Photonic Technology for Integrated Sensing and Communication,Ministry of Education,and Guangdong Provincial Key Laboratory of Information Photonics Technology,Institute of Advanced Photonics Technology,School of Information Engineering,Guangdong University of Technology,Guangzhou 510006,China;Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai),Zhuhai 519082,China;Key Laboratory of Light Field Manipulation and Information Acquisition,Ministry of Industry and Information Technology,and Shaanxi Key Laboratory of Optical Information Technology,School of Physical Science and Technology,Northwestern Polytechnical University,Xi′an 710129,China)

机构地区：[1]广东工业大学信息工程学院通感融合光子技术教育部重点实验室广东省信息光子技术重点实验室先进光子技术研究院,广州510006 [2]南方海洋科学与工程广东省实验室(珠海),珠海519082 [3]西北工业大学物理科学与技术学院光场调控与信息感知工业和信息化部重点实验室陕西省光信息技术重点实验室,西安710129

出　　处：《光子学报》2024年第8期247-259,共13页Acta Photonica Sinica

基　　金：国家自然科学基金(Nos.62075183,62275218);广东省“珠江人才计划”引进创新创业团队(Nos.2021ZT09X044,2019ZT08X340);中央高校基本科研业务费专项资金(No.D5000230117)。

摘　　要：针对光学遥感图像中微小目标空间分辨率低、有效特征不足等问题,在YOLOv5检测算法基础上,提出一种基于多尺度分层残差网络的光学遥感图像微小目标检测方法。设计了一种简单高效的多尺度分层残差特征提取模块,可在更细粒度水平上获得更丰富的感受野,强化神经网络的特征提取能力,进一步提升微小目标特征丰富度。在此基础上,进一步优化损失函数中的定位损失项,通过增加距离惩罚提升检测算法对微小目标的定位能力。在光学遥感微小目标检测数据集AI-TODv2和微小行人检测数据集TinyPerson上开展了系统对比实验,实验结果表明所提出算法相较于基准YOLOv5算法平均精度分别提升了5.5%和1.8%,有效提高了微小目标检测的召回率和准确率。Optical remote sensing image object detection aims to precisely locate and categorize targets such as aircraft,vehicles,and ships.Challenges arise due to the vast distances in remote sensing,leading to numerous tiny objects that are hard to characterize.Additionally,complex backgrounds and environmental factors like lighting and weather conditions reduce signal-to-noise ratios,increasing detection difficulties.Although Convolutional Neural Networks(CNNs),e specially t hose f rom t he Y OLO f amily,are employed for their efficient feature extraction capabilities,they perform poorly in detecting these tiny objects.The key to realize the detection of tiny objects in optical remote sensing images is to obtain sufficiently rich multi-scale feature information and clear tiny object features.Aiming at the above problems,this paper proposes a multi-scale hierarchical residual network based optical remote sensing image tiny object detection algorithm MHRM-YOLO o n t he b asis o f Y OLOv5,and designs a simple and efficient Multi-scale Hierarchical Residual tiny object feature extraction Module(MHRM).This module expands on Cross Stage Partial(CSP)module by doing more layered design and using different convolutional combinations to extract features from different layered,which allows the network to obtain richer gradient information flow and output richer feature map combinations.In addition,MHRM can be easily embedded into the existing mainstream YOLO detection algorithm backbone network,which can obtain richer sensory fields at a finer granularity level and can effectively capture the contextual information of tiny objects and retain their spatial feature information.The network structure of the MHRM-YOLO algorithm is mainly divided into three parts,namely the backbone,the neck,and the head for prediction.The backbone consists of MHRM and basic convolution module,which performs finegrained feature extraction to obtain more multi-scale information and larger sensory field;the neck part uses the conventional CSP plus Path Ag

关 键 词：光学遥感图像 微小目标检测 深度学习 多尺度 卷积神经网络 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]