Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing Images
Semantic segmentation is a key technology for remote sensing image analysis widely used in land cover classification, natural disaster monitoring, and other fields. Unlike traditional image segmentation, there are various targets in remote sensing images, with a large feature difference between the...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/6/1347 |
| _version_ | 1797612171510677504 |
|---|---|
| author | Jintong Jia Jiarui Song Qingqiang Kong Huan Yang Yunhe Teng Xuan Song |
| author_facet | Jintong Jia Jiarui Song Qingqiang Kong Huan Yang Yunhe Teng Xuan Song |
| author_sort | Jintong Jia |
| collection | DOAJ |
| description | Semantic segmentation is a key technology for remote sensing image analysis widely used in land cover classification, natural disaster monitoring, and other fields. Unlike traditional image segmentation, there are various targets in remote sensing images, with a large feature difference between the targets. As a result, segmentation is more difficult, and the existing models retain low accuracy and inaccurate edge segmentation when used in remote sensing images. This paper proposes a multi-attention-based semantic segmentation network for remote sensing images in order to address these problems. Specifically, we choose UNet as the baseline model, using a coordinate attention-based residual network in the encoder to improve the extraction capability of the backbone network for fine-grained features. We use a content-aware reorganization module in the decoder to replace the traditional upsampling operator to improve the network information extraction capability, and, in addition, we propose a fused attention module for feature map fusion after upsampling, aiming to solve the multi-scale problem. We evaluate our proposed model on the WHDLD dataset and our self-labeled Lu County dataset. The model achieved an mIOU of 63.27% and 72.83%, and an mPA of 74.86% and 84.72%, respectively. Through comparison and confusion matrix analysis, our model outperformed commonly used benchmark models on both datasets. |
| first_indexed | 2024-03-11T06:38:32Z |
| format | Article |
| id | doaj.art-175310e269af4a71aa57db2ae459cf84 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-11T06:38:32Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-175310e269af4a71aa57db2ae459cf842023-11-17T10:44:14ZengMDPI AGElectronics2079-92922023-03-01126134710.3390/electronics12061347Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing ImagesJintong Jia0Jiarui Song1Qingqiang Kong2Huan Yang3Yunhe Teng4Xuan Song5School of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450001, ChinaDepartment of Mathematics, University of Toronto, Toronto, ON M5S 1A1, CanadaThe First Geological Survey Exploration Institute of Henan Bureau of Geo-Exploration and Mineral Development, Zhengzhou 450001, ChinaSchool of Water Conservancy Science and Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Cyber Science and Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSemantic segmentation is a key technology for remote sensing image analysis widely used in land cover classification, natural disaster monitoring, and other fields. Unlike traditional image segmentation, there are various targets in remote sensing images, with a large feature difference between the targets. As a result, segmentation is more difficult, and the existing models retain low accuracy and inaccurate edge segmentation when used in remote sensing images. This paper proposes a multi-attention-based semantic segmentation network for remote sensing images in order to address these problems. Specifically, we choose UNet as the baseline model, using a coordinate attention-based residual network in the encoder to improve the extraction capability of the backbone network for fine-grained features. We use a content-aware reorganization module in the decoder to replace the traditional upsampling operator to improve the network information extraction capability, and, in addition, we propose a fused attention module for feature map fusion after upsampling, aiming to solve the multi-scale problem. We evaluate our proposed model on the WHDLD dataset and our self-labeled Lu County dataset. The model achieved an mIOU of 63.27% and 72.83%, and an mPA of 74.86% and 84.72%, respectively. Through comparison and confusion matrix analysis, our model outperformed commonly used benchmark models on both datasets.https://www.mdpi.com/2079-9292/12/6/1347remote sensing imageattention mechanismimage segmentationdeep learningsemantic segmentation |
| spellingShingle | Jintong Jia Jiarui Song Qingqiang Kong Huan Yang Yunhe Teng Xuan Song Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing Images Electronics remote sensing image attention mechanism image segmentation deep learning semantic segmentation |
| title | Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing Images |
| title_full | Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing Images |
| title_fullStr | Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing Images |
| title_full_unstemmed | Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing Images |
| title_short | Multi-Attention-Based Semantic Segmentation Network for Land Cover Remote Sensing Images |
| title_sort | multi attention based semantic segmentation network for land cover remote sensing images |
| topic | remote sensing image attention mechanism image segmentation deep learning semantic segmentation |
| url | https://www.mdpi.com/2079-9292/12/6/1347 |
| work_keys_str_mv | AT jintongjia multiattentionbasedsemanticsegmentationnetworkforlandcoverremotesensingimages AT jiaruisong multiattentionbasedsemanticsegmentationnetworkforlandcoverremotesensingimages AT qingqiangkong multiattentionbasedsemanticsegmentationnetworkforlandcoverremotesensingimages AT huanyang multiattentionbasedsemanticsegmentationnetworkforlandcoverremotesensingimages AT yunheteng multiattentionbasedsemanticsegmentationnetworkforlandcoverremotesensingimages AT xuansong multiattentionbasedsemanticsegmentationnetworkforlandcoverremotesensingimages |