4RATFNet: Four-Dimensional Residual-Attention Improved-Transfer Few-Shot Semantic Segmentation Network for Landslide Detection

Landslides are hazardous and in many cases can cause enormous economic losses and human casualties. The suddenness of landslides makes it difficult to detect landslides quickly and effectively. Therefore, to address the problem of intelligent analysis of geological landslides, we propose a 4RATFNet network for few-shot semantic segmentation detection in the case of insufficient number of labeled landslide images. First, a residual-attention module is designed to fuse channel features and spatial features for residual fusion. Second, improved transfer learning is used to optimize the parameters of the pre-trained network. Third, the network downscales the four-dimensional convolutional kernel into a pair of two-dimensional convolutional kernels. Finally, the few-shot semantic segmentation network is used to extract support image features and complete the landslide detection for the same features in the query image. The experimental results show that the method performs better when tested on Resnet50 backbone and Resnet101 backbone when the sample size of labeled landslide images is insufficient. Compared with traditional semantic segmentation methods, it can obtain better segmentation results and achieve higher mean intersection over union, indicating that our network has obvious advantages and wider applicability.