Abstract:Landslide disasters pose a significant threat, with their highly destructive nature underscoring the critical importance of timely and accurate recognition for effective early warning systems and emergency response efforts. In recent years, substantial advancements have been made in the realm of landslide recognition (LR) based on remote sensing data, leveraging deep learning techniques. However, the intricate and varied environments in which landslides occur often present challenges in detecting subtle changes, especially when relying solely on optical remote sensing images. InSAR (Interferometric Synthetic Aperture Radar) technology emerges as a valuable tool for LR, providing more detailed ground deformation data and enhancing the theoretical foundation. To harness the slow deformation characteristics of landslides, we developed the FCADenseNet model. This model is designed to learn features and patterns within ground deformation data, with a specific focus on improving LR. A noteworthy aspect of our model is the integration of an attention mechanism, which considers various monitoring factors. This holistic approach enables the comprehensive detection of landslide disasters across entire watersheds, providing valuable information on landslide hazards. Our experimental results demonstrate the effectiveness of the FCADenseNet model, with an F1-score of 0.7611, which is 9.53% higher than that of FC_DenseNet. This study substantiates the feasibility and efficacy of combining InSAR with deep learning methods for LR. The insights gained from this research contribute to the advancement of regional landslide geological hazard monitoring, identification, and prevention strategies.

What problem does this paper attempt to address?

The paper attempts to address the problem of efficiently and accurately identifying landslide disasters in complex and variable environments. Specifically, the paper focuses on the following points: 1. **Challenges of Landslide Identification**: Traditional landslide identification methods (such as manual visual interpretation) are highly subjective and time-consuming, making it difficult to meet the demands of processing large amounts of information. Although computer-aided methods and mainstream machine learning algorithms (such as artificial neural networks, support vector machines, random forests, etc.) have improved performance to some extent, they still face difficulties in handling large-scale remote sensing data and extracting deep-level information. 2. **Environmental Complexity**: The environment in which landslides occur is usually very complex, with significant variations in terrain, vegetation, and land use types. Additionally, adverse weather conditions (especially cloud cover) affect the quality and usability of optical remote sensing images, making it difficult to capture surface changes. 3. **Limitations of Existing Technologies**: Although deep learning-based methods have made significant progress in landslide identification, relying solely on optical remote sensing images still makes it difficult to detect subtle changes, especially in complex environments. To address the above issues, the paper proposes a new model that combines InSAR technology and deep learning methods—FCADenseNet. This model aims to improve the accuracy and reliability of landslide identification by integrating surface deformation data and multi-source information. Specifically, the main contributions of the paper include: - **Introduction of Attention Mechanism**: By introducing attention modules in the skip connections of the encoder and decoder, the model better utilizes low-level fine-grained information and high-level semantic features, avoiding information loss and edge blurring. - **Multi-Source Data Fusion**: By combining optical remote sensing images, geological factor data, and surface deformation information obtained from InSAR, the model establishes regional landslide sample data and explores the coupling effects between these factors. - **Model Performance Validation**: Experimental results show that the FCADenseNet model achieves an F1 score of 0.7611 in the landslide identification task, which is 9.53% higher than the FC_DenseNet model, demonstrating the effectiveness of the proposed model. Overall, the paper provides a more comprehensive and accurate landslide identification solution by combining InSAR technology and deep learning methods, contributing to the improvement of geological disaster monitoring, identification, and prevention capabilities.

Deep Learning-Based Landslide Recognition Incorporating Deformation Characteristics

A Deep-Learning-Based Algorithm for Landslide Detection over Wide Areas Using InSAR Images Considering Topographic Features

Automatic recognition of active landslides by surface deformation and deep learning

Landslide Extraction Using Mask R-CNN with Background-Enhancement Method

A New Deep-Learning-Based Approach for Earthquake-Triggered Landslide Detection From Single-Temporal RapidEye Satellite Imagery

Landslide Hazard Identification Based on Deep Learning and Sentinel-2 Remote Sensing Imagery

Landslide Recognition Based on Machine Learning Considering Terrain Feature Fusion

Landslide Recognition by Deep Convolutional Neural Network and Change Detection

Landslide identification using machine learning

Accurate landslide identification by multisource data fusion analysis with improved feature extraction backbone network

Deep Fusion of Local and Non-Local Features for Precision Landslide Recognition

Landslide detection based on shipborne images and deep learning models: a case study in the Three Gorges Reservoir Area in China

Utilizing deep learning approach to develop landslide susceptibility mapping considering landslide types

Automatic and Efficient Detection of Loess Landslides Based on Deep Learning

A Novel Method for Identifying Landslide Surface Deformation via the Integrated YOLOX and Mask R-CNN Model

Landslide Susceptibility Prediction Modeling Based on Remote Sensing and a Novel Deep Learning Algorithm of a Cascade-Parallel Recurrent Neural Network

Study of the Automatic Recognition of Landslides by Using InSAR Images and the Improved Mask R-CNN Model in the Eastern Tibet Plateau

Automatic Remote Sensing Identification of Co-Seismic Landslides Using Deep Learning Methods

An integrated neural network method for landslide susceptibility assessment based on time-series InSAR deformation dynamic features

Remote Sensing Landslide Recognition Based on Convolutional Neural Network

Robust Landslide Recognition Using UAV Datasets: A Case Study in Baihetan Reservoir