A5073858307- Landslides and related hazards
- Fire effects on ecosystems
- Cryospheric studies and observations
- Tree Root and Stability Studies
- Flood Risk Assessment and Management
- Advanced Computational Techniques and Applications
- Energy and Environment Impacts
- Anomaly Detection Techniques and Applications
- Solar Radiation and Photovoltaics
- Urban Heat Island Mitigation
Chengdu University
2023-2025
Chengdu University of Technology
2023-2024
Landslide susceptibility assessment is crucial for preventing landslide risks. However, existing methods only consider local environmental features related to landslides, neglecting remote yet interconnected geographical features, leading unreliable maps. This study fully considers the complex terrain and landform of mountainous areas where landslides occur. From perspectives mapping units models, it introduces correlations achieve a comprehensive association between affected environments,...
Quantification of shading effects from complex terrain on solar radiation is essential to obtain precise data incident in mountainous areas. In this study, a machine learning (ML) approach proposed rapidly estimate the radiation. Based two different ML algorithms, namely, Ordinary Least Squares (OLS) and Gradient Boosting Decision Tree (GBDT), uses terrain-related factors as input variables model analyze direct diffuse rates. case study western Sichuan, annual rates were most correlated with...
Artificial Intelligence (AI) technologies have profoundly transformed the field of remote sensing, revolutionizing data collection, processing, and analysis. Traditionally reliant on manual interpretation task-specific models, sensing has been significantly enhanced by advent foundation models--large-scale, pre-trained AI models capable performing a wide array tasks with unprecedented accuracy efficiency. This paper provides comprehensive survey in domain, covering released between June 2021...
Considering the great time and labor consumption involved in conventional hazard assessment methods compiling landslide inventory, construction of a transferable susceptibility prediction model is crucial. This study employs UAV images as data sources to interpret typical alpine valley area Beichuan County. Eight environmental factors including digital elevation (DEM) are extracted establish pixel-wise dataset, along with interpreted data. Two models were built, each deep neural network...
Landslides are most severe in the mountainous regions of southwestern China. While landslide identification provides a foundation for disaster prevention operations, methods utilizing multi-source data and deep learning techniques to improve efficiency accuracy complex environments still focus research difficult issue research. In this study, we address above problems construct model based on shifted window (Swin) transformer. We chose Ya’an, which has terrain experiences frequent...
Landslide susceptibility prediction usually involves the comprehensive analysis of terrain and other factors that may be distributed with spatial patterns. Without considering correlation mutual influence between pixels, conventional methods often focus only on information from individual pixels. To address this issue, present study proposes a new strategy for neighboring pixel collaboration based Unified Perceptual Parsing Network (UPerNet), Vision Transformer (ViT), Graph Neural Networks...