A landslide extraction method of channel attention mechanism U-Net network based on Sentinel-2A remote sensing images. Issue 1 (31st December 2023)

• Record Type: Journal Article
• Authors: Chen, Hesheng; He, Yi; Zhang, Lifeng; Yao, Sheng; Yang, Wang; Fang, Yumin; Liu, Yaoxiang; Gao, Binghai
• Is Part Of: International journal of digital earth. Volume 16:Issue 1(2023)
• Page Start: 552
• Page End: 577
• Publication Date: 2023-12-31
• Subjects: Sentinel-2A remote sensing image -- landslide extraction -- U-Net -- attention mechanism -- deep learning; Geographic information systems -- Periodicals; Sustainable development -- Information technology -- Periodicals; Social planning -- Information technology -- Periodicals; 910.285
• DOI: 10.1080/17538947.2023.2177359
• Languages: English
• ISSNs: 1753-8947

Abstract

ABSTRACT: Accurate landslide extraction is significant for landslide disaster prevention and control. Remote sensing images have been widely used in landslide investigation, and landslide extraction methods based on deep learning combined with remote sensing images (such as U-Net) have received a lot of attention. However, because of the variable shape and texture features of landslides in remote sensing images, the rich spectral features, and the complexity of their surrounding features, landslide extraction using U-Net can lead to problems such as false detection and missed detection. Therefore, this study introduces the channel attention mechanism called the squeeze-and-excitation network (SENet) in the feature fusion part of U-Net; the study also constructs an attention U-Net landside extraction model combining SENet and U-Net, and uses Sentinel-2A remote sensing images for model training and validation. The extraction results are evaluated through different evaluation metrics and compared with those of two models: U-Net and U-Net Backbone (U-Net Without Skip Connection). The results show that proposed the model can effectively extract landslides based on Sentinel-2A remote sensing images with an F1 value of 87.94%, which is about 2% and 3% higher than U-Net and U-Net Backbone, respectively, with less false detection and more accurate extraction results.