SAR Target Classification Based on Sample Spectral Regularization
Synthetic Aperture Radar (SAR) target classification is an important branch of SAR image interpretation. The deep learning based SAR target classification algorithms have made remarkable achievements. But the acquisition and annotation of SAR target images are time-consuming and laborious, and it is...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-11-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/12/21/3628 |
id |
doaj-f67d21af5d6647cf944b380ad7e1b8d7 |
---|---|
record_format |
Article |
spelling |
doaj-f67d21af5d6647cf944b380ad7e1b8d72020-11-25T04:03:26ZengMDPI AGRemote Sensing2072-42922020-11-01123628362810.3390/rs12213628SAR Target Classification Based on Sample Spectral RegularizationWei Liang0Tengfei Zhang1Wenhui Diao2Xian Sun3Liangjin Zhao4Kun Fu5Yirong Wu6Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaSynthetic Aperture Radar (SAR) target classification is an important branch of SAR image interpretation. The deep learning based SAR target classification algorithms have made remarkable achievements. But the acquisition and annotation of SAR target images are time-consuming and laborious, and it is difficult to obtain sufficient training data in many cases. The insufficient training data can make deep learning based models suffering from over-fitting, which will severely limit their wide application in SAR target classification. Motivated by the above problem, this paper employs transfer-learning to transfer the prior knowledge learned from a simulated SAR dataset to a real SAR dataset. To overcome the sample restriction problem caused by the poor feature discriminability for real SAR data. A simple and effective sample spectral regularization method is proposed, which can regularize the singular values of each SAR image feature to improve the feature discriminability. Based on the proposed regularization method, we design a transfer-learning pipeline to leverage the simulated SAR data as well as acquire better feature discriminability. The experimental results indicate that the proposed method is feasible for the sample restriction problem in SAR target classification. Furthermore, the proposed method can improve the classification accuracy when relatively sufficient training data is available, and it can be plugged into any convolutional neural network (CNN) based SAR classification models.https://www.mdpi.com/2072-4292/12/21/3628SAR target classification, transfer-learning, spectral regularization |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Wei Liang Tengfei Zhang Wenhui Diao Xian Sun Liangjin Zhao Kun Fu Yirong Wu |
spellingShingle |
Wei Liang Tengfei Zhang Wenhui Diao Xian Sun Liangjin Zhao Kun Fu Yirong Wu SAR Target Classification Based on Sample Spectral Regularization Remote Sensing SAR target classification, transfer-learning, spectral regularization |
author_facet |
Wei Liang Tengfei Zhang Wenhui Diao Xian Sun Liangjin Zhao Kun Fu Yirong Wu |
author_sort |
Wei Liang |
title |
SAR Target Classification Based on Sample Spectral Regularization |
title_short |
SAR Target Classification Based on Sample Spectral Regularization |
title_full |
SAR Target Classification Based on Sample Spectral Regularization |
title_fullStr |
SAR Target Classification Based on Sample Spectral Regularization |
title_full_unstemmed |
SAR Target Classification Based on Sample Spectral Regularization |
title_sort |
sar target classification based on sample spectral regularization |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2020-11-01 |
description |
Synthetic Aperture Radar (SAR) target classification is an important branch of SAR image interpretation. The deep learning based SAR target classification algorithms have made remarkable achievements. But the acquisition and annotation of SAR target images are time-consuming and laborious, and it is difficult to obtain sufficient training data in many cases. The insufficient training data can make deep learning based models suffering from over-fitting, which will severely limit their wide application in SAR target classification. Motivated by the above problem, this paper employs transfer-learning to transfer the prior knowledge learned from a simulated SAR dataset to a real SAR dataset. To overcome the sample restriction problem caused by the poor feature discriminability for real SAR data. A simple and effective sample spectral regularization method is proposed, which can regularize the singular values of each SAR image feature to improve the feature discriminability. Based on the proposed regularization method, we design a transfer-learning pipeline to leverage the simulated SAR data as well as acquire better feature discriminability. The experimental results indicate that the proposed method is feasible for the sample restriction problem in SAR target classification. Furthermore, the proposed method can improve the classification accuracy when relatively sufficient training data is available, and it can be plugged into any convolutional neural network (CNN) based SAR classification models. |
topic |
SAR target classification, transfer-learning, spectral regularization |
url |
https://www.mdpi.com/2072-4292/12/21/3628 |
work_keys_str_mv |
AT weiliang sartargetclassificationbasedonsamplespectralregularization AT tengfeizhang sartargetclassificationbasedonsamplespectralregularization AT wenhuidiao sartargetclassificationbasedonsamplespectralregularization AT xiansun sartargetclassificationbasedonsamplespectralregularization AT liangjinzhao sartargetclassificationbasedonsamplespectralregularization AT kunfu sartargetclassificationbasedonsamplespectralregularization AT yirongwu sartargetclassificationbasedonsamplespectralregularization |
_version_ |
1724440159292751872 |