N-Cluster Loss and Hard Sample Generative Deep Metric Learning for PolSAR Image Classification

• Published in: IEEE Transactions on Geoscience and Remote Sensing
• Main Authors: Yang, Chen, Hou, Biao, Chanussot, Jocelyn, Hu, Yue, Ren, Bo, Wang, Shuang, Jiao, Licheng
• Other Authors: Xidian University, GIPSA - Signal Images Physique (GIPSA-SIGMAPHY), Observatoire des Sciences de l'Univers de Grenoble (OSUG )-GIPSA Pôle Sciences des Données (GIPSA-PSD), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA)-Observatoire des Sciences de l'Univers de Grenoble (OSUG ), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut national des sciences de l'Univers (INSU - CNRS)-Institut national de recherche en sciences et technologies pour l'environnement et l'agriculture (IRSTEA)-Université Savoie Mont Blanc (USMB Université de Savoie Université de Chambéry )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes 2016-2019 (UGA 2016-2019 )-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut national des sciences de l'Univers (INSU - CNRS)-Institut national de recherche en sciences et technologies pour l'environnement et l'agriculture (IRSTEA)-Université Savoie Mont Blanc (USMB Université de Savoie Université de Chambéry )-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes 2016-2019 (UGA 2016-2019 ), Apprentissage de modèles à partir de données massives (Thoth), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire Jean Kuntzmann (LJK), Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)
• Format: Article in Journal/Newspaper
• Language: English
• Published: HAL CCSD 2022
• Subjects: Measurement; Task analysis; Feature extraction; Scattering; Training; Data mining; Convergence; [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing; DML
• Online Access: https://hal.science/hal-03932552; https://doi.org/10.1109/TGRS.2021.3099840

International audience Deep learning works normally in PolSAR image classification because the complex terrain scattering characteristic results in large intraclass differences and high interclass similarity. Deep metric learning (DML) aims to make the features keep a closer intraclass and a farther interclass distance. Therefore, we introduce DML and then propose an N-cluster generative adversarial net (N-cluster GAN) framework for PolSAR image classification. However, existing DML losses mainly focus on the relationship between individual samples in feature space. Hence, we propose N-cluster loss that pays more attention to the overall structure of all samples. Meanwhile, traditional hard negative sample mining methods occupy lots of computational resources. In addition, the hard level of the negative samples will affect the model’s performance. Therefore, we explore a new method based on a GAN framework to replace the sample mining. Positive N-cluster loss is added to the discriminator ( D ), and a negative one is added to the generator ( G ). In this way, D will possess better classification ability, and G can produce hard negative samples for D . Then, the hard level of the generated negative samples will change with the discrimination of D , which is appropriate for the proposed model. N-cluster loss can be directly calculated through the extracted features rather than redundant data preparation. The proposed model is verified on four PolSAR datasets from two aspects of the loss function and negative samples mining. Then, it achieves competitive performance compared with state-of-the-art algorithms.