Land Cover Classification for Polarimetric SAR Images Based on Vision Transformer
Open Access
- 18 September 2022
- journal article
- research article
- Published by MDPI AG in Remote Sensing
- Vol. 14 (18) , 4656
Abstract
Deep learning methods have been widely studied for Polarimetric synthetic aperture radar (PolSAR) land cover classification. The scarcity of PolSAR labeled samples and the small receptive field of the model limit the performance of deep learning methods for land cover classification. In this paper, a vision Transformer (ViT)-based classification method is proposed. The ViT structure can extract features from the global range of images based on a self-attention block. The powerful feature representation capability of the model is equivalent to a flexible receptive field, which is suitable for PolSAR image classification at different resolutions. In addition, because of the lack of labeled data, the Mask Autoencoder method is used to pre-train the proposed model with unlabeled data. Experiments are carried out on the Flevoland dataset acquired by NASA/JPL AIRSAR and the Hainan dataset acquired by the Aerial Remote Sensing System of the Chinese Academy of Sciences. The experimental results on both datasets demonstrate the superiority of the proposed method.Keywords
Funding Information
- Major Project of Chinese High-resolution Earth Observation System (30-H30C01-9004-19/21, 62171023, 62222102)
- National Natural Science Foundation of China (30-H30C01-9004-19/21, 62171023, 62222102)
This publication has 37 references indexed in Scilit:
- Classification of Polarimetric SAR Images Based on Modeling Contextual Information and Using Texture FeaturesIEEE Transactions on Geoscience and Remote Sensing, 2015
- The WGΓ Distribution for Multilook Polarimetric SAR Data and Its ApplicationIEEE Geoscience and Remote Sensing Letters, 2015
- Deep learningNature, 2015
- Land Cover Classification for Polarimetric SAR Images Based on Mixture ModelsRemote Sensing, 2014
- Four-Component Scattering Power Decomposition With Rotation of Coherency MatrixIEEE Transactions on Geoscience and Remote Sensing, 2011
- The polarimetric 𝒢 distribution for SAR data analysisEnvironmetrics, 2004
- A three-component scattering model for polarimetric SAR dataIEEE Transactions on Geoscience and Remote Sensing, 1998
- An entropy based classification scheme for land applications of polarimetric SARIEEE Transactions on Geoscience and Remote Sensing, 1997
- Classification of multi-look polarimetric SAR imagery based on complex Wishart distributionInternational Journal of Remote Sensing, 1994
- Classification of Earth terrain using polarimetric synthetic aperture radar imagesJournal of Geophysical Research, 1989