PolSAR Image Land Cover Classification Based on Hierarchical Capsule Network
Polarimetric synthetic aperture <span style="font-variant: small-caps;">r</span>adar (PolSAR) image classification is one of the basic methods of PolSAR image interpretation. Deep learning algorithms, especially convolutional neural networks (CNNs), have been widely used in Pol...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-08-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/16/3132 |
| _version_ | 1797522162750324736 |
|---|---|
| author | Jianda Cheng Fan Zhang Deliang Xiang Qiang Yin Yongsheng Zhou Wei Wang |
| author_facet | Jianda Cheng Fan Zhang Deliang Xiang Qiang Yin Yongsheng Zhou Wei Wang |
| author_sort | Jianda Cheng |
| collection | DOAJ |
| description | Polarimetric synthetic aperture <span style="font-variant: small-caps;">r</span>adar (PolSAR) image classification is one of the basic methods of PolSAR image interpretation. Deep learning algorithms, especially convolutional neural networks (CNNs), have been widely used in PolSAR image classification due to their powerful feature learning capabilities. However, a single neuron in the CNN cannot represent multiple polarimetric attributes of the land cover. The capsule network (CapsNet) uses vectors instead of the single neuron to characterize the polarimetric attributes, which improves the classification performance compared with traditional CNNs. In this paper, a hierarchical capsule network (HCapsNet) is proposed for the land cover classification of PolSAR images, which can consider the deep features obtained at different network levels in the classification. Moreover, we adopt three attributes to uniformly describe the scattering mechanisms of different land covers: phase, amplitude, and polarimetric decomposition parameters, which improves the generalization performance of HCapsNet. Furthermore, conditional random field (CRF) is added to the classification framework to eliminate small isolated regions of the intra-class. Comprehensive evaluations are performed on three PolSAR datasets acquired by different sensors, which demonstrate that our proposed method outperforms other state-of-the-art methods. |
| first_indexed | 2024-03-10T08:25:34Z |
| format | Article |
| id | doaj.art-98acc0114ce04daa84671318a50f99fe |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T08:25:34Z |
| publishDate | 2021-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-98acc0114ce04daa84671318a50f99fe2023-11-22T09:32:27ZengMDPI AGRemote Sensing2072-42922021-08-011316313210.3390/rs13163132PolSAR Image Land Cover Classification Based on Hierarchical Capsule NetworkJianda Cheng0Fan Zhang1Deliang Xiang2Qiang Yin3Yongsheng Zhou4Wei Wang5College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, ChinaCollege of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, ChinaBeijing Advanced Innovation Center for Soft Matter Science and Engineering, Beijing University of Chemical Technology, Beijing 100029, ChinaCollege of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, ChinaCollege of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, ChinaCollege of Electronic Science and Technology, National University of Defense Technology, Changsha 410073, ChinaPolarimetric synthetic aperture <span style="font-variant: small-caps;">r</span>adar (PolSAR) image classification is one of the basic methods of PolSAR image interpretation. Deep learning algorithms, especially convolutional neural networks (CNNs), have been widely used in PolSAR image classification due to their powerful feature learning capabilities. However, a single neuron in the CNN cannot represent multiple polarimetric attributes of the land cover. The capsule network (CapsNet) uses vectors instead of the single neuron to characterize the polarimetric attributes, which improves the classification performance compared with traditional CNNs. In this paper, a hierarchical capsule network (HCapsNet) is proposed for the land cover classification of PolSAR images, which can consider the deep features obtained at different network levels in the classification. Moreover, we adopt three attributes to uniformly describe the scattering mechanisms of different land covers: phase, amplitude, and polarimetric decomposition parameters, which improves the generalization performance of HCapsNet. Furthermore, conditional random field (CRF) is added to the classification framework to eliminate small isolated regions of the intra-class. Comprehensive evaluations are performed on three PolSAR datasets acquired by different sensors, which demonstrate that our proposed method outperforms other state-of-the-art methods.https://www.mdpi.com/2072-4292/13/16/3132land cover classificationPolSAR imageHCapsNetCRF |
| spellingShingle | Jianda Cheng Fan Zhang Deliang Xiang Qiang Yin Yongsheng Zhou Wei Wang PolSAR Image Land Cover Classification Based on Hierarchical Capsule Network Remote Sensing land cover classification PolSAR image HCapsNet CRF |
| title | PolSAR Image Land Cover Classification Based on Hierarchical Capsule Network |
| title_full | PolSAR Image Land Cover Classification Based on Hierarchical Capsule Network |
| title_fullStr | PolSAR Image Land Cover Classification Based on Hierarchical Capsule Network |
| title_full_unstemmed | PolSAR Image Land Cover Classification Based on Hierarchical Capsule Network |
| title_short | PolSAR Image Land Cover Classification Based on Hierarchical Capsule Network |
| title_sort | polsar image land cover classification based on hierarchical capsule network |
| topic | land cover classification PolSAR image HCapsNet CRF |
| url | https://www.mdpi.com/2072-4292/13/16/3132 |
| work_keys_str_mv | AT jiandacheng polsarimagelandcoverclassificationbasedonhierarchicalcapsulenetwork AT fanzhang polsarimagelandcoverclassificationbasedonhierarchicalcapsulenetwork AT deliangxiang polsarimagelandcoverclassificationbasedonhierarchicalcapsulenetwork AT qiangyin polsarimagelandcoverclassificationbasedonhierarchicalcapsulenetwork AT yongshengzhou polsarimagelandcoverclassificationbasedonhierarchicalcapsulenetwork AT weiwang polsarimagelandcoverclassificationbasedonhierarchicalcapsulenetwork |