面向遥感图像目标感知的群目标检测框架  

A group object detection framework for remote sensing image object perception

作　　者：张鸿伟 金磊[1] 邹学超 方宇强 尹璐[4] 赵健 兴军亮 ZHANG Hongwei;JIN Lei;ZOU Xuechao;FANG Yuqiang;YIN Lu;ZHAO Jian;XING Junliang(School of Electronic Engineering,Beijing University of Posts and Telecommunications,Beijing 100876,China;School of Computer Technology and Applications,Qinghai University,Qinghai 810016,China;Graduate School,University of Aerospace Engineering,Beijing 101416,China;Beijing Institute of Remote Sensing Information,Beijing 100192,China;North Institute of Electronic Equipment,Beijing 100071,China;School of Computing,Tsinghua University,Beijing 100084,China)

机构地区：[1]北京邮电大学电子工程学院,北京100876 [2]青海大学计算机技术与应用学院,西宁810016 [3]航天工程大学研究生院,北京101416 [4]北京遥感信息研究所,北京100192 [5]北方电子设备研究所,北京100071 [6]清华大学计算机学院,北京100084

出　　处：《遥感学报》2024年第7期1802-1811,共10页NATIONAL REMOTE SENSING BULLETIN

基　　金：卫星信息智能处理与应用技术实验室基金(编号:2022-ZZKY-JJ-06-01)。

摘　　要：光学遥感是航天侦察和地质勘测中的常用技术,拍摄得到的可见光图像能够提供非常丰富的信息,在目标监视、态势预判等方面都具有重要应用。近年来以轮船、飞机等物体检测为代表的光学遥感图像目标感知取得了显著进展,但对于目标尺度变化大,目标数量多而小的遥感图像目标感知场景中还存在巨大技术挑战,也就是在当前的光学遥感图像目标感知场景存在很多目标小并且多目标集中的情况,容易导致误检和漏检。为了解决现有遥感图像目标检测算法不同目标独立检测的内在低效性,本文提出了一种新的检测框架,即群目标检测,以期通过检测群目标的状态信息来缓解单一目标感知信息不足、可靠性差等问题,进而得到更为可靠的多目标检测结果。本文首先对群目标的概念进行定义,然后基于该定义提出了一种群目标自动化标注方案,在公开数据集上对原有标签进行分析,无需任何手动标注,就能得到含有群目标标注的注释信息。基于群目标自动化标注,本文提出了群目标检测算法,即在检测群目标的同时,利用群目标的空间约束提升单一目标检测结果。实验证明,与近年来的遥感图像检测算法相比,本文提出的群目标检测在最热门的大型遥感目标检测数据集DOTA上验证时,性能最佳。Optical remote sensing is a widely used technology in aerospace reconnaissance and geological exploration.Visible light images captured by this technology provide a wealth of information and have important applications in intelligence gathering,object monitoring,and situational forecasting.Considerable progress in remote sensing image object perception has been achieved,particularly in ship and airplane detection.However,technical challenges,including with large object-scale variations and numerous small objects,in remote sensing image object perception remain.Existing work has mainly focused on improving boundary box representations,and single-object detection models fail to fully exploit spatial correlation information from surrounding or similar objects.To address the inherent inefficiency of existing remote sensing image object detection algorithms that detect different objects independently,this paper proposes a novel detection framework called group object detection.By detecting the state information of a group object,our framework alleviates problems,such as insufficient perception information and poor reliability of single-object perception,generating reliable multi-object detection results.This paper introduces a concept of group objects and proposes an automated annotation scheme for group objects.By analyzing existing labels on a public dataset,the proposed scheme obtains annotated information with group object labels without manual annotation.Based on the automated annotation of group targets,a group target detection algorithm is presented,which enhances single-object detection results by utilizing the spatial constraints of group objects.Experimental results on the DOTA dataset,a widely-used remote sensing object detection benchmark,demonstrate that the proposed group target detection algorithm outperforms state-of-the-art methods.

关 键 词：遥感图像 目标检测 边界框 群目标 自动化标注 DOTA 目标感知 多目标 

分 类 号：P2[天文地球—测绘科学与技术]