Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep Learning
Accurate monitoring of bare soil land (BSL) is an urgent need for environmental governance and optimal utilization of land resources. High-resolution imagery contains rich semantic information, which is beneficial for the recognition of objects on the ground. Simultaneously, it is susceptible to the...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/6/1646 |
| _version_ | 1827747823286747136 |
|---|---|
| author | Chen He Yalan Liu Dacheng Wang Shufu Liu Linjun Yu Yuhuan Ren |
| author_facet | Chen He Yalan Liu Dacheng Wang Shufu Liu Linjun Yu Yuhuan Ren |
| author_sort | Chen He |
| collection | DOAJ |
| description | Accurate monitoring of bare soil land (BSL) is an urgent need for environmental governance and optimal utilization of land resources. High-resolution imagery contains rich semantic information, which is beneficial for the recognition of objects on the ground. Simultaneously, it is susceptible to the impact of its background. We propose a semantic segmentation model, Deeplabv3+-M-CBAM, for extracting BSL. First, we replaced the Xception of Deeplabv3+ with MobileNetV2 as the backbone network to reduce the number of parameters. Second, to distinguish BSL from the background, we employed the convolutional block attention module (CBAM) via a combination of channel attention and spatial attention. For model training, we built a BSL dataset based on BJ-2 satellite images. The test result for the F1 of the model was 88.42%. Compared with Deeplabv3+, the classification accuracy improved by 8.52%, and the segmentation speed was 2.34 times faster. In addition, compared with the visual interpretation, the extraction speed improved by 11.5 times. In order to verify the transferable performance of the model, Jilin-1GXA images were used for the transfer test, and the extraction accuracies for F1, IoU, recall and precision were 86.07%, 87.88%, 87.00% and 95.80%, respectively. All of these experiments show that Deeplabv3+-M-CBAM achieved efficient and accurate extraction results and a well transferable performance for BSL. The methodology proposed in this study exhibits its application value for the refinement of environmental governance and the surveillance of land use. |
| first_indexed | 2024-03-11T05:57:41Z |
| format | Article |
| id | doaj.art-d994629f1ba149409e3f2e2f0de844ad |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T05:57:41Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-d994629f1ba149409e3f2e2f0de844ad2023-11-17T13:40:00ZengMDPI AGRemote Sensing2072-42922023-03-01156164610.3390/rs15061646Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep LearningChen He0Yalan Liu1Dacheng Wang2Shufu Liu3Linjun Yu4Yuhuan Ren5Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, ChinaAccurate monitoring of bare soil land (BSL) is an urgent need for environmental governance and optimal utilization of land resources. High-resolution imagery contains rich semantic information, which is beneficial for the recognition of objects on the ground. Simultaneously, it is susceptible to the impact of its background. We propose a semantic segmentation model, Deeplabv3+-M-CBAM, for extracting BSL. First, we replaced the Xception of Deeplabv3+ with MobileNetV2 as the backbone network to reduce the number of parameters. Second, to distinguish BSL from the background, we employed the convolutional block attention module (CBAM) via a combination of channel attention and spatial attention. For model training, we built a BSL dataset based on BJ-2 satellite images. The test result for the F1 of the model was 88.42%. Compared with Deeplabv3+, the classification accuracy improved by 8.52%, and the segmentation speed was 2.34 times faster. In addition, compared with the visual interpretation, the extraction speed improved by 11.5 times. In order to verify the transferable performance of the model, Jilin-1GXA images were used for the transfer test, and the extraction accuracies for F1, IoU, recall and precision were 86.07%, 87.88%, 87.00% and 95.80%, respectively. All of these experiments show that Deeplabv3+-M-CBAM achieved efficient and accurate extraction results and a well transferable performance for BSL. The methodology proposed in this study exhibits its application value for the refinement of environmental governance and the surveillance of land use.https://www.mdpi.com/2072-4292/15/6/1646bare soil landhigh-resolution remote sensing imagerysemantic segmentationdeep learningDeeplabv3+CBAM |
| spellingShingle | Chen He Yalan Liu Dacheng Wang Shufu Liu Linjun Yu Yuhuan Ren Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep Learning Remote Sensing bare soil land high-resolution remote sensing imagery semantic segmentation deep learning Deeplabv3+ CBAM |
| title | Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep Learning |
| title_full | Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep Learning |
| title_fullStr | Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep Learning |
| title_full_unstemmed | Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep Learning |
| title_short | Automatic Extraction of Bare Soil Land from High-Resolution Remote Sensing Images Based on Semantic Segmentation with Deep Learning |
| title_sort | automatic extraction of bare soil land from high resolution remote sensing images based on semantic segmentation with deep learning |
| topic | bare soil land high-resolution remote sensing imagery semantic segmentation deep learning Deeplabv3+ CBAM |
| url | https://www.mdpi.com/2072-4292/15/6/1646 |
| work_keys_str_mv | AT chenhe automaticextractionofbaresoillandfromhighresolutionremotesensingimagesbasedonsemanticsegmentationwithdeeplearning AT yalanliu automaticextractionofbaresoillandfromhighresolutionremotesensingimagesbasedonsemanticsegmentationwithdeeplearning AT dachengwang automaticextractionofbaresoillandfromhighresolutionremotesensingimagesbasedonsemanticsegmentationwithdeeplearning AT shufuliu automaticextractionofbaresoillandfromhighresolutionremotesensingimagesbasedonsemanticsegmentationwithdeeplearning AT linjunyu automaticextractionofbaresoillandfromhighresolutionremotesensingimagesbasedonsemanticsegmentationwithdeeplearning AT yuhuanren automaticextractionofbaresoillandfromhighresolutionremotesensingimagesbasedonsemanticsegmentationwithdeeplearning |