Dual-branch PolSAR Image Classification Based on GraphMAE and Local Feature Extraction
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916350487691264 |
|---|---|
| author | Wang, Yuchen Guo, Ziyi Bi, Haixia Hong, Danfeng Xu, Chen |
| author_facet | Wang, Yuchen Guo, Ziyi Bi, Haixia Hong, Danfeng Xu, Chen |
| contents | The annotation of polarimetric synthetic aperture radar (PolSAR) images is a labor-intensive and time-consuming process. Therefore, classifying PolSAR images with limited labels is a challenging task in remote sensing domain. In recent years, self-supervised learning approaches have proven effective in PolSAR image classification with sparse labels. However, we observe a lack of research on generative selfsupervised learning in the studied task. Motivated by this, we propose a dual-branch classification model based on generative self-supervised learning in this paper. The first branch is a superpixel-branch, which learns superpixel-level polarimetric representations using a generative self-supervised graph masked autoencoder. To acquire finer classification results, a convolutional neural networks-based pixel-branch is further incorporated to learn pixel-level features. Classification with fused dual-branch features is finally performed to obtain the predictions. Experimental results on the benchmark Flevoland dataset demonstrate that our approach yields promising classification results. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_04294 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Dual-branch PolSAR Image Classification Based on GraphMAE and Local Feature Extraction Wang, Yuchen Guo, Ziyi Bi, Haixia Hong, Danfeng Xu, Chen Computer Vision and Pattern Recognition Machine Learning The annotation of polarimetric synthetic aperture radar (PolSAR) images is a labor-intensive and time-consuming process. Therefore, classifying PolSAR images with limited labels is a challenging task in remote sensing domain. In recent years, self-supervised learning approaches have proven effective in PolSAR image classification with sparse labels. However, we observe a lack of research on generative selfsupervised learning in the studied task. Motivated by this, we propose a dual-branch classification model based on generative self-supervised learning in this paper. The first branch is a superpixel-branch, which learns superpixel-level polarimetric representations using a generative self-supervised graph masked autoencoder. To acquire finer classification results, a convolutional neural networks-based pixel-branch is further incorporated to learn pixel-level features. Classification with fused dual-branch features is finally performed to obtain the predictions. Experimental results on the benchmark Flevoland dataset demonstrate that our approach yields promising classification results. |
| title | Dual-branch PolSAR Image Classification Based on GraphMAE and Local Feature Extraction |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2408.04294 |