计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 240100058-9.doi: 10.11896/jsjkx.240100058

• 图像处理&多媒体技术 • 上一篇    下一篇

基于双重标签分配的遥感有向目标检测方法

董燕1,2, 魏铭宏1, 高广帅1, 刘洲峰1, 李春雷1   

  1. 1 中原工学院信息与通信工程学院 郑州 450007
    2 电子科技大学自动化工程学院 成都 610000
  • 出版日期:2024-11-16 发布日期:2024-11-13
  • 通讯作者: 李春雷(lichunlei1979@zut.edu.cn)
  • 作者简介:(dy@zut.edu.cn)
  • 基金资助:
    国家自然科学基金(62072489);中原科技创新领军人才项目(234200510009);河南省科技攻关项目(232102211002,232102211030)

Remote Sensing Orineted Object Detection Method Based on Dual-label Assignment

DONG Yan1,2, WEI Minghong1, GAO Guangshuai1, LIU Zhoufeng1, LI Chunlei1   

  1. 1 School of Information and Communication Engineering,Zhongyuan University of Technology,Zhengzhou 450007,China
    2 School of Automation Engineering,University of Electronic Science and Technology,Chengdu 610000,China
  • Online:2024-11-16 Published:2024-11-13
  • About author:DONG Yan,born in 1977,Ph.D,professor.Her main research interests include artificial intelligence,pattern recognition and surface defect detection based on machine vision.
    LI Chunlei,born in 1979,Ph.D,professor.His main research interests include computer vision and pattern recognition.
  • Supported by:
    NSFC(62072489),Leading Talents of Science and Technology in the Central Plain of China(234200510009) and Henan Province Key Science and Technology Research Projects(232102211002,232102211030).

PDF (PC)

179

摘要： 由于遥感图像目标具有任意方向、大纵横比和密集排列等多样性分布特点,预设的锚框难以精准匹配所有真实目标,导致对大纵横比和密集排列的有向目标检测精度不高。为了解决上述问题,提出了一种基于双重标签分配的遥感有向目标检测方法。首先,提出双重标签分配策略为目标分配最大及次优交并比的候选框;其次,通过排斥损失(AP-Loss)和吸引损失(UP-Loss)约束相邻目标的候选框,以提高目标正确匹配概率;然后,为了提取适应于分类和回归分支的鲁棒特征,设计了一个特征增强模块(FEM),该模块基于偏振函数构造自适应特征,能够有效增强分类和回归任务所需的特征表达能力;最后,设计了一个定位指导分类(LGC)模块,该模块通过定位任务指导分类任务的采样位置,进行定位细化,以获取分类任务的关键特征,从而缓解分类与定位之间的不一致问题。在3个公开的遥感有向目标检测数据集DOTA,HRSC-2016和DIOR-R上进行了大量的实验,实验结果证明了所提方法的有效性,且优于现有主流方法。

关键词: 遥感图像, 有向目标检测, 双重标签分配, 不一致问题, 采样位置细化

Abstract: Due to the inherently diverse distribution characteristics of remote sensing image objects,such as their arbitrary orientation,large aspect ratio,and densely arranged,the utilization of preset anchor boxes makes it difficult to accurately match all real objects.This limitation results in low detection accuracy,particularly for oriented objects with large aspect ratios and densely arranged.To address this issue,an oriented object detection method in remote sensing images based on dual label assignment is proposed.Firstly,a dual-label assignment strategy is proposed to assign candidate boxes with maximum and suboptimal intersection union ratios to the real object.Then,the candidate boxes of adjacent objects are constrained by repulsion loss(AP Loss) and attraction loss(UP Loss) to improve the probability of correct object matching.In addition,to extract robust features suitable for classification and regression branches,a Feature Enhancement Module(FEM) is designed.This module constructs adaptive features based on polarization functions,which can effectively enhance the feature expression ability required for classification and regression tasks.Finally,a localization-guided classification(LGC) module is designed,which guides the sampling position of the classification task through localization tasks,performs localization refinement,and obtains key features of the classification task,thereby alleviating the inconsistency between classification and localization.A large number of experiments were conducted on three publicly available oriented object detection in remote sensing datasets,namely DOTA,HRSC-2016,and DIOR-R.The experimental results demonstrated the effectiveness of the proposed method and its accuracy(mAP) is better than existing mainstream methods.

Key words: Remote sensing images, Oriented object detection, Dual-label assignment, Inconsistency issues, Sampling position refinement

中图分类号: 

  • TP751
[1]HAN J,DING J,LI J,et al.Align deep features for oriented object detection[J].IEEE Transactions on Geoscience and Remote Sensing,2021,60:1-11.
[2]XIA G S,BAI X,DING J,et al.DOTA:A large-scale dataset for object detection in aerial images[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:3974-3983.
[3]RAN Y,ZHANG L.R-YOLOv5:Auto-cutting,R-otated TextDetection Model[J].Computer Science,2022,49(S2):637-642.
[4]QIAN W,YANG X,PENG S,et al.RSDet++:Point-basedmodulated loss for more accurate rotated object detection[J].IEEE Transactions on Circuits and Systems for Video Technology,2022,32(11):7869-7879.
[5]YANG X,YAN J,FENG Z,et al.R3det:Refined single-stagedetector with feature refinement for rotating object[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021,35(4):3163-3171.
[6]LI X,WANG J G.Optimization Algorithm for Detection and Localization of Pure Face in Color Images[J].Computer Science,2009,36(7):284-287.
[7]WANG W,CHENG J W,WANG X.Remote Sensing Targets Detection Based on Adaptive Weighting Feature Dictionaries and Joint Sparse[J].Computer Science,2018,45(10):276-280.
[8]ZHU Y,FANG G S,ZHENG B B,et al.Research on DetectionMethod of Refined Rotated Boxes in Remote Sensing[J].Acta Automatica Sinica,2023,49(2):415-424.
[9]LU Q,YU Y Q,XU D M,et al.Improved YOLOv5 for Small Drones Target Detection Algorithm[J].Computer Science,2023,50(S2):212-219.
[10]MA J,SHAO W,YE H,et al.Arbitrary-oriented scene text detection via rotation proposals[J].IEEE Transactions on Multimedia,2018,20:3111-3122.
[11]DING J,XUE N,LONG Y,et al.Learning roi transformer fororiented object detection in aerial images[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).2019:2844-2853.
[12]GUO Z,ZHANG X,LIU C,et al.Convex-hull feature adapta-tion for oriented and densely packed object detection[J].IEEE Transactions on Circuits and Systems for Video Technology,2022,32:5252-5265.
[13]XU Y,FU M,WANG Q,et al.Gliding vertex on the horizontal bounding box for multi-oriented object detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,43:1452-1459.
[14]ZHANG S,CHI C,YAO Y,et al.Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).2020:9756-9765.
[15]MING Q,ZHOU Z,MIAO L,et al.Dynamic anchor learning for arbitrary-oriented object detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021:2355-2363.
[16]YANG X,YAN J,MING Q,et al.Rethinking rotated object detection wi5th gaussian wasserstein distance loss[C]//International Conference on Machine Learning.PMLR,2021:11830-11841.
[17]WANG J,XU C,YANG W,et al.A normalized gaussian wasserstein distance for tiny object detection[J].arXiv,vol.abs/2110.13389,2021.
[18]HUANG Z,LI W,XIA X G,et al.A general gaussian heatmap label assignment for arbitrary-oriented object detection[J].IEEE Transactions on Image Processing,2022,31:1895-1910.
[19]REN S,HE K,GIRSHICK R B,et al.Faster r-cnn:Towardsreal-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39:1137-1149.
[20]LIN T Y,GOYAL P,GIRSHICK R B,et al.Focal loss for dense object detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2020,42:318-327.
[21]CHENG B,WEI Y,SHI H,et al.Revisiting rcnn:On awakening the classification power of faster rcnn[C]//Proceedings of the European Conference on Computer Vision(ECCV).2018:453-468.
[22]CHENG B,WEI Y,FERIS R,et al.Decoupled classification refinement:Hard false positive suppression for object detection[J].arXiv:1810.04002,2018.
[23]FENG C,ZHONG Y,GAO Y,et al.Tood:Task-aligned one-stage object detection[C]//2021 IEEE/CVF International Conference on Computer Vision(ICCV).2021:3490-3499.
[24]ZHANG S,WEN L,LEI Z,et al.Refinedet++:Single-shot refinement neural network for object detection[J].IEEE Transactions on Circuits and Systems for Video Technology,2021,31:674-687.
[25]MING Q,MIAO L,ZHOU Z,et al.CFC-Net:A Critical Feature Capturing Network for Arbitrary-Oriented Object Detection in Remote-Sensing Images[C]//IEEE Transactions on Geoscience and Remote Sensing,2022,60:1-14.
[26]LIU Z,YUAN L,WENG L,et al.A high resolution optical satellite image dataset for ship recognition and some new baselines[C]//International Conference on Pattern Recognition Applications and Methods.2017.
[27]CHENG G,WANG J,LI K,et al.Anchor-free oriented proposal generator for object detection[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:1-11.
[28]CHENG G,YAO Y,LI S,et al.Dual-aligned oriented detector[J].IEEE Transactions on Geoscience and Remote Sensing,2022,60:1-11.
[29]WANG X,XIAO T,JIANG Y,et al.Repulsion Loss:Detecting Pedestrians in a Crowd[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City,UT,USA,2018:7774-7783.
[30]ZHANG G,LU S,ZHANG W.CAD-Net:A context-aware detection network for objects in remote sensing imagery[J].IEEE Trans.Geosci.Remote Sens.,2019,57(12):10015-10024.
[31]YANG X,YANG J,YAN J,et al.SCRDet:Towards more robust detection for small,cluttered and rotated objects[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:8232-8241.
[32]PAN X,REN Y,SHENG K,et al.Dynamic refinement network for oriented and densely packed object detection[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).2020:11204-11213.
[33]CHEN Z,CHEN K,LIN W,et al. Piou loss:Towards accurate oriented object detection in complex environments[C]//Computer Vision-ECCV 2020:16th European Conference,Glasgow,UK,August 23-28,2020,Proceedings,Part V 16.Springer International Publishing,2020:195-211.
[34]QIAN W,YANG X,PENG S,et al.Learning modulated loss for rotated object detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021:2458-2466.
[35]XIE X,CHENG G,WANG J,et al.Oriented r-cnn for object de-tection[C]//2021 IEEE/CVF International Conference on Computer Vision(ICCV).2021:3500-3509.
[36]LI W,CHEN Y,HU K,et al.Oriented reppoints for aerial object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:1829-1838.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发