The development of artificial intelligence makes it possible to rapidly segment landslides. However, there are still some challenges in landslide segmentation based on remote sensing images, such as low segmentation accuracy, caused by similar features, inhomogeneous features, and blurred boundaries. To address these issues, we propose a novel deep learning model called AST-UNet in this paper. This model is based on structure of SwinUNet, attaching a channel Attention and spatial intersection (CASI) module as a parallel branch of the encoder, and a spatial detail enhancement (SDE) module in the skip connection. Specifically, (1) the spatial intersection module expands the spatial attention range, alleviating noise in the image and enhances the continuity of landslides in segmentation results; (2) the channel attention module refines the spatial attention weights by feature modeling in the channel dimension, improving the model’s ability to differentiate targets that closely resemble landslides; and (3) the spatial detail enhancement module increases the accuracy for landslide boundaries by strengthening the attention of the decoder to detailed features. We use the landslide data from the area of Luding, Sichuan to conduct experiments. The comparative analyses with state-of-the-art (SOTA) models, including FCN, UNet, DeepLab V3+, TransFuse, TranUNet, and SwinUNet, prove the superiority of our AST-UNet for landslide segmentation. The generalization of our model is also verified in the experiments. The proposed AST-UNet obtains an F1-score of 90.14%, mIoU of 83.45%, foreground IoU of 70.81%, and Hausdorff distance of 3.73, respectively, on the experimental datasets.

The China–Pakistan Economic Corridor is the pilot area of the Belt and Road, where glaciers and lakes are widely distributed. Recent years, global warming has accelerated the expansion of glacier lakes, which increased the risk of natural disasters such as glacier lake outburst. It is important to monitor the glacier lakes in this region. In this paper, we propose a method combining the object-oriented image analysis with boundary recognition (OOBR) to extract lakes in several study areas of China–Pakistan Economic Corridor (CPEC). This method recognized the lake boundary with the symmetrical characteristic according to the principle of seed growth of watershed algorithm, which can correct the boundary extracted by the object-oriented method. The overall accuracy of the proposed method is up to 98.5% with Landsat series images. The experiments also show that the overall accuracy of our method is always higher than that of the object-oriented method with different segmentation scales mentioned in this paper. The proposed method improved the overall accuracy on the basis of the results obtained by the object-oriented method, and the results with the proposed method are more robust to the seeds than that with the boundary correction method of the watershed algorithm. Therefore, the proposed method can obtain a high extraction accuracy while reducing the complexity of the object-oriented extraction.