Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification
Deep-learning-based SAR ship classification has become a research hotspot in the military and civilian fields and achieved remarkable performance. However, the volume of available SAR ship classification data is relatively small, meaning that previous deep-learning-based methods have usually struggl...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-08-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/15/16/4058 |
_version_ | 1797583343459500032 |
---|---|
author | Congan Xu Long Gao Hang Su Jianting Zhang Junfeng Wu Wenjun Yan |
author_facet | Congan Xu Long Gao Hang Su Jianting Zhang Junfeng Wu Wenjun Yan |
author_sort | Congan Xu |
collection | DOAJ |
description | Deep-learning-based SAR ship classification has become a research hotspot in the military and civilian fields and achieved remarkable performance. However, the volume of available SAR ship classification data is relatively small, meaning that previous deep-learning-based methods have usually struggled with overfitting problems. Moreover, due to the limitation of the SAR imaging mechanism, the large intraclass diversity and small interclass similarity further degrade the classification performance. To address these issues, we propose a label smoothing auxiliary classifier generative adversarial network with triplet loss (LST-ACGAN) for SAR ship classification. In our method, an ACGAN is introduced to generate SAR ship samples with category labels. To address the model collapse problem in the ACGAN, the smooth category labels are assigned to generated samples. Moreover, triplet loss is integrated into the ACGAN for discriminative feature learning to enhance the margin of different classes. Extensive experiments on the OpenSARShip dataset demonstrate the superior performance of our method compared to the previous methods. |
first_indexed | 2024-03-10T23:36:35Z |
format | Article |
id | doaj.art-5c16dbd88b824e7094b3da315d959fb4 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-10T23:36:35Z |
publishDate | 2023-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-5c16dbd88b824e7094b3da315d959fb42023-11-19T02:53:57ZengMDPI AGRemote Sensing2072-42922023-08-011516405810.3390/rs15164058Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship ClassificationCongan Xu0Long Gao1Hang Su2Jianting Zhang3Junfeng Wu4Wenjun Yan5Advanced Technology Research Institute, Beijing Institute of Technology, Jinan 250300, ChinaInformation Fusion Institute, Naval Aviation University, Yantai 264000, ChinaInformation Fusion Institute, Naval Aviation University, Yantai 264000, ChinaNo. 91977 Unit of People’s Liberation Army of China, Beijing 100036, ChinaInformation Fusion Institute, Naval Aviation University, Yantai 264000, ChinaInformation Fusion Institute, Naval Aviation University, Yantai 264000, ChinaDeep-learning-based SAR ship classification has become a research hotspot in the military and civilian fields and achieved remarkable performance. However, the volume of available SAR ship classification data is relatively small, meaning that previous deep-learning-based methods have usually struggled with overfitting problems. Moreover, due to the limitation of the SAR imaging mechanism, the large intraclass diversity and small interclass similarity further degrade the classification performance. To address these issues, we propose a label smoothing auxiliary classifier generative adversarial network with triplet loss (LST-ACGAN) for SAR ship classification. In our method, an ACGAN is introduced to generate SAR ship samples with category labels. To address the model collapse problem in the ACGAN, the smooth category labels are assigned to generated samples. Moreover, triplet loss is integrated into the ACGAN for discriminative feature learning to enhance the margin of different classes. Extensive experiments on the OpenSARShip dataset demonstrate the superior performance of our method compared to the previous methods.https://www.mdpi.com/2072-4292/15/16/4058SAR ship classificationgenerative adversarial networkslabel smoothingdeep metric learning |
spellingShingle | Congan Xu Long Gao Hang Su Jianting Zhang Junfeng Wu Wenjun Yan Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification Remote Sensing SAR ship classification generative adversarial networks label smoothing deep metric learning |
title | Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification |
title_full | Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification |
title_fullStr | Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification |
title_full_unstemmed | Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification |
title_short | Label Smoothing Auxiliary Classifier Generative Adversarial Network with Triplet Loss for SAR Ship Classification |
title_sort | label smoothing auxiliary classifier generative adversarial network with triplet loss for sar ship classification |
topic | SAR ship classification generative adversarial networks label smoothing deep metric learning |
url | https://www.mdpi.com/2072-4292/15/16/4058 |
work_keys_str_mv | AT conganxu labelsmoothingauxiliaryclassifiergenerativeadversarialnetworkwithtripletlossforsarshipclassification AT longgao labelsmoothingauxiliaryclassifiergenerativeadversarialnetworkwithtripletlossforsarshipclassification AT hangsu labelsmoothingauxiliaryclassifiergenerativeadversarialnetworkwithtripletlossforsarshipclassification AT jiantingzhang labelsmoothingauxiliaryclassifiergenerativeadversarialnetworkwithtripletlossforsarshipclassification AT junfengwu labelsmoothingauxiliaryclassifiergenerativeadversarialnetworkwithtripletlossforsarshipclassification AT wenjunyan labelsmoothingauxiliaryclassifiergenerativeadversarialnetworkwithtripletlossforsarshipclassification |