Exploring Reinforced Class Separability and Discriminative Representations for SAR Target Open Set Recognition

被引:1
|
作者
Gao, Fei [1 ]
Luo, Xin [1 ]
Lang, Rongling [1 ]
Wang, Jun [1 ]
Sun, Jinping [1 ]
Hussain, Amir [2 ]
机构
[1] Beihang Univ, Sch Elect & Informat Engn, Beijing 100191, Peoples R China
[2] Edinburgh Napier Univ, Cyber & Big Data Res Lab, Edinburgh EH11 4BN, Scotland
基金
英国工程与自然科学研究理事会; 中国国家自然科学基金;
关键词
synthetic aperture radar (SAR); open set recognition (OSR); class separability reinforcement; discriminative feature capture; embedding space optimization; CLASSIFICATION;
D O I
10.3390/rs16173277
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Current synthetic aperture radar (SAR) automatic target recognition (ATR) algorithms primarily operate under the closed-set assumption, implying that all target classes have been previously learned during the training phase. However, in open scenarios, they may encounter target classes absent from the training set, thereby necessitating an open set recognition (OSR) challenge for SAR-ATR. The crux of OSR lies in establishing distinct decision boundaries between known and unknown classes to mitigate confusion among different classes. To address this issue, we introduce a novel framework termed reinforced class separability for SAR target open set recognition (RCS-OSR), which focuses on optimizing prototype distribution and enhancing the discriminability of features. First, to capture discriminative features, a cross-modal causal features enhancement module (CMCFE) is proposed to strengthen the expression of causal regions. Subsequently, regularized intra-class compactness loss (RIC-Loss) and intra-class relationship aware consistency loss (IRC-Loss) are devised to optimize the embedding space. In conjunction with joint supervised training using cross-entropy loss, RCS-OSR can effectively reduce empirical classification risk and open space risk simultaneously. Moreover, a class-aware OSR classifier with adaptive thresholding is designed to leverage the differences between different classes. Consequently, our method can construct distinct decision boundaries between known and unknown classes to simultaneously classify known classes and identify unknown classes in open scenarios. Extensive experiments conducted on the MSTAR dataset demonstrate the effectiveness and superiority of our method in various OSR tasks.
引用
收藏
页数:25
相关论文
共 50 条
  • [41] Open Set Audio Recognition for Multi-Class Classification With Rejection
    Jleed, Hitham
    Bouchard, Martin
    IEEE ACCESS, 2020, 8 : 146523 - 146534
  • [42] Multi-class Open Set Recognition Using Probability of Inclusion
    Jain, Lalit P.
    Scheirer, Walter J.
    Boult, Terrance E.
    COMPUTER VISION - ECCV 2014, PT III, 2014, 8691 : 393 - 409
  • [43] Towards Discriminative Representations and Unbiased Predictions: Class-specific Angular Softmax for Speech Emotion Recognition
    Li, Zhixuan
    He, Liang
    Li, Jingyang
    Wang, Li
    Zhang, Wei-Qiang
    INTERSPEECH 2019, 2019, : 1696 - 1700
  • [44] An Open-Set Recognition Approach for SAR Targets Using Only Classification Scores
    Sun, Qian
    Chen, Shichao
    Wu, Lirong
    Su, Jia
    Tao, Mingliang
    Liu, Ming
    2024 4TH URSI ATLANTIC RADIO SCIENCE MEETING, AT-RASC 2024, 2024,
  • [45] Proportional Similarity-Based Openmax Classifier for Open Set Recognition in SAR Images
    Giusti, Elisa
    Ghio, Selenia
    Oveis, Amir Hosein
    Martorella, Marco
    REMOTE SENSING, 2022, 14 (18)
  • [46] Joint sparse representation for multi-resolution representations of SAR images with application to target recognition
    Zhang, Zhenyu
    Liu, Sikai
    JOURNAL OF ELECTROMAGNETIC WAVES AND APPLICATIONS, 2018, 32 (11) : 1342 - 1353
  • [47] Vehicle Make and Model Recognition as an Open-Set Recognition Problem and New Class Discovery
    Vazquez-Santiago, Diana-Itzel
    Acosta-Mesa, Hector-Gabriel
    Mezura-Montes, Efren
    MATHEMATICAL AND COMPUTATIONAL APPLICATIONS, 2023, 28 (04)
  • [48] Discriminative Feature Learning for Real-Time SAR Automatic Target Recognition With the Nonlinear Analysis Cosparse Model
    Wen, Zaidao
    Hou, Biao
    Wu, Qian
    Jiao, Licheng
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2018, 15 (07) : 1045 - 1049
  • [49] Radar HRRP Open Set Target Recognition Based on Closed Classification Boundary
    Xia, Ziheng
    Wang, Penghui
    Liu, Hongwei
    REMOTE SENSING, 2023, 15 (02)
  • [50] SYNTHETIC MINORITY CLASS DATA BY GENERATIVE ADVERSARIAL NETWORK FOR IMBALANCED SAR TARGET RECOGNITION
    Luo, Zhongming
    Jiang, Xue
    Liu, Xingzhao
    IGARSS 2020 - 2020 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2020, : 2459 - 2462