A Siamese Swin-Unet for image change detection
Abstract The problem of change detection in remote sensing image processing is both difficult and important. It is extensively used in a variety of sectors, including land resource planning, monitoring and forecasting of agricultural plant health, and monitoring and assessment of natural disasters....
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-02-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-54096-8 |
| _version_ | 1797274918582222848 |
|---|---|
| author | Yizhuo Tang Zhengtao Cao Ningbo Guo Mingyong Jiang |
| author_facet | Yizhuo Tang Zhengtao Cao Ningbo Guo Mingyong Jiang |
| author_sort | Yizhuo Tang |
| collection | DOAJ |
| description | Abstract The problem of change detection in remote sensing image processing is both difficult and important. It is extensively used in a variety of sectors, including land resource planning, monitoring and forecasting of agricultural plant health, and monitoring and assessment of natural disasters. Remote sensing images provide a large amount of long-term and fully covered data for earth environmental monitoring. A lot of progress has been made thanks to deep learning's quick development. But the majority of deep learning-based change detection techniques currently in use rely on the well-known Convolutional neural network (CNN). However, considering the locality of convolutional operation, CNN unable to master the interplay between global and distant semantic information. Some researches has employ Vision Transformer as a backbone in remote sensing field. Inspired by these researches, in this paper, we propose a network named Siam-Swin-Unet, which is a Siamesed pure Transformer with U-shape construction for remote sensing image change detection. Swin Transformer is a hierarchical vision transformer with shifted windows that can extract global feature. To learn local and global semantic feature information, the dual-time image are fed into Siam-Swin-Unet which is composed of Swin Transformer, Unet Siamesenet and two feature fusion module. Considered the Unet and Siamesenet are effective for change detection, We applied it to the model. The feature fusion module is designed for fusion of dual-time image features, and is efficient and low-compute confirmed by our experiments. Our network achieved 94.67 F1 on the CDD dataset (season varying). |
| first_indexed | 2024-03-07T15:05:59Z |
| format | Article |
| id | doaj.art-29420b36d89d4dbe970337c19920043e |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-07T15:05:59Z |
| publishDate | 2024-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-29420b36d89d4dbe970337c19920043e2024-03-05T18:52:47ZengNature PortfolioScientific Reports2045-23222024-02-011411910.1038/s41598-024-54096-8A Siamese Swin-Unet for image change detectionYizhuo Tang0Zhengtao Cao1Ningbo Guo2Mingyong Jiang3Space Engineering UniversitySpace Engineering UniversitySpace Engineering UniversitySpace Engineering UniversityAbstract The problem of change detection in remote sensing image processing is both difficult and important. It is extensively used in a variety of sectors, including land resource planning, monitoring and forecasting of agricultural plant health, and monitoring and assessment of natural disasters. Remote sensing images provide a large amount of long-term and fully covered data for earth environmental monitoring. A lot of progress has been made thanks to deep learning's quick development. But the majority of deep learning-based change detection techniques currently in use rely on the well-known Convolutional neural network (CNN). However, considering the locality of convolutional operation, CNN unable to master the interplay between global and distant semantic information. Some researches has employ Vision Transformer as a backbone in remote sensing field. Inspired by these researches, in this paper, we propose a network named Siam-Swin-Unet, which is a Siamesed pure Transformer with U-shape construction for remote sensing image change detection. Swin Transformer is a hierarchical vision transformer with shifted windows that can extract global feature. To learn local and global semantic feature information, the dual-time image are fed into Siam-Swin-Unet which is composed of Swin Transformer, Unet Siamesenet and two feature fusion module. Considered the Unet and Siamesenet are effective for change detection, We applied it to the model. The feature fusion module is designed for fusion of dual-time image features, and is efficient and low-compute confirmed by our experiments. Our network achieved 94.67 F1 on the CDD dataset (season varying).https://doi.org/10.1038/s41598-024-54096-8Change detectionRemote sensingSwin transformerSwin-UnetSiamesenet |
| spellingShingle | Yizhuo Tang Zhengtao Cao Ningbo Guo Mingyong Jiang A Siamese Swin-Unet for image change detection Scientific Reports Change detection Remote sensing Swin transformer Swin-Unet Siamesenet |
| title | A Siamese Swin-Unet for image change detection |
| title_full | A Siamese Swin-Unet for image change detection |
| title_fullStr | A Siamese Swin-Unet for image change detection |
| title_full_unstemmed | A Siamese Swin-Unet for image change detection |
| title_short | A Siamese Swin-Unet for image change detection |
| title_sort | siamese swin unet for image change detection |
| topic | Change detection Remote sensing Swin transformer Swin-Unet Siamesenet |
| url | https://doi.org/10.1038/s41598-024-54096-8 |
| work_keys_str_mv | AT yizhuotang asiameseswinunetforimagechangedetection AT zhengtaocao asiameseswinunetforimagechangedetection AT ningboguo asiameseswinunetforimagechangedetection AT mingyongjiang asiameseswinunetforimagechangedetection AT yizhuotang siameseswinunetforimagechangedetection AT zhengtaocao siameseswinunetforimagechangedetection AT ningboguo siameseswinunetforimagechangedetection AT mingyongjiang siameseswinunetforimagechangedetection |