A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab Model
Due to the various types of land cover and large spectral differences in remote sensing images, high-quality semantic segmentation of these images still faces challenges such as fuzzy object boundary extraction and difficulty in identifying small targets. To address these challenges, this study prop...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Land |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-445X/14/1/22 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588225645903872 |
|---|---|
| author | Jinxin Wang Manman Wang Kaiwei Cong Zilong Qin |
| author_facet | Jinxin Wang Manman Wang Kaiwei Cong Zilong Qin |
| author_sort | Jinxin Wang |
| collection | DOAJ |
| description | Due to the various types of land cover and large spectral differences in remote sensing images, high-quality semantic segmentation of these images still faces challenges such as fuzzy object boundary extraction and difficulty in identifying small targets. To address these challenges, this study proposes a new improved model based on the TransDeepLab segmentation method. The model introduces a GAM attention mechanism in the coding stage, and incorporates a multi-level linear up-sampling strategy in the decoding stage. These enhancements allow the model to fully utilize multi-level semantic information and small target details in high-resolution remote sensing images, thereby effectively improving the segmentation accuracy of target objects. Using the open-source LoveDA large remote sensing image datasets for the validation experiment, the results show that compared to the original model, the improved model’s MIOU increased by 2.68%, aACC by 3.41%, and mACC by 4.65%. Compared to other mainstream models, the model also achieved superior segmentation performance. |
| format | Article |
| id | doaj-art-b113c69ad95140f884b9d1f7d567aac3 |
| institution | Kabale University |
| issn | 2073-445X |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Land |
| spelling | doaj-art-b113c69ad95140f884b9d1f7d567aac32025-01-24T13:37:35ZengMDPI AGLand2073-445X2024-12-011412210.3390/land14010022A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab ModelJinxin Wang0Manman Wang1Kaiwei Cong2Zilong Qin3School of Geoscience & Technology, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Geoscience & Technology, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Computer and Information Technology, Northeast Petroleum University, Daqing 163000, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, ChinaDue to the various types of land cover and large spectral differences in remote sensing images, high-quality semantic segmentation of these images still faces challenges such as fuzzy object boundary extraction and difficulty in identifying small targets. To address these challenges, this study proposes a new improved model based on the TransDeepLab segmentation method. The model introduces a GAM attention mechanism in the coding stage, and incorporates a multi-level linear up-sampling strategy in the decoding stage. These enhancements allow the model to fully utilize multi-level semantic information and small target details in high-resolution remote sensing images, thereby effectively improving the segmentation accuracy of target objects. Using the open-source LoveDA large remote sensing image datasets for the validation experiment, the results show that compared to the original model, the improved model’s MIOU increased by 2.68%, aACC by 3.41%, and mACC by 4.65%. Compared to other mainstream models, the model also achieved superior segmentation performance.https://www.mdpi.com/2073-445X/14/1/22deep learninghigh-resolution remote sensing imagesemantic segmentationfeature extraction |
| spellingShingle | Jinxin Wang Manman Wang Kaiwei Cong Zilong Qin A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab Model Land deep learning high-resolution remote sensing image semantic segmentation feature extraction |
| title | A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab Model |
| title_full | A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab Model |
| title_fullStr | A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab Model |
| title_full_unstemmed | A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab Model |
| title_short | A Semantic Segmentation Method for Remote Sensing Images Based on an Improved TransDeepLab Model |
| title_sort | semantic segmentation method for remote sensing images based on an improved transdeeplab model |
| topic | deep learning high-resolution remote sensing image semantic segmentation feature extraction |
| url | https://www.mdpi.com/2073-445X/14/1/22 |
| work_keys_str_mv | AT jinxinwang asemanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel AT manmanwang asemanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel AT kaiweicong asemanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel AT zilongqin asemanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel AT jinxinwang semanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel AT manmanwang semanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel AT kaiweicong semanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel AT zilongqin semanticsegmentationmethodforremotesensingimagesbasedonanimprovedtransdeeplabmodel |