Landslide Susceptibility Mapping Using an Image–Tabular Joint Deep Learning Framework: A Case Study of the Tacheng Region, Xinjiang, China
Qianjie Deng, Dingfan Xing, Xiong Wu, L. SONG, ZhuoEr TENG, Rui Wang, Shicheng Gao, Zhiwu Zhang, Kun‐Feng Qiu
Accurate landslide susceptibility mapping (LSM) is important for hazard prevention and land use planning in mountainous regions. Existing machine learning and deep learning methods mainly use raster-based conditioning factors. They often ignore landslide-related attribute information and spatial context. To address this issue, this study proposes an image–tabular joint deep learning framework for regional-scale LSM. The framework is based on a FiLM-conditioned U-Net. The model combines raster patches with an estimated soft attribute-prior vector and uses FiLM to guide condition-aware spatial feature learning. The proposed framework was tested in the Tacheng region, Xinjiang, China. The dataset includes a landslide inventory and conditioning factors related to terrain, hydrology, vegetation, geology, land cover, and human activities. Model performance was evaluated using stratified five-fold cross-validation, an independent test set, buffer-radius sensitivity tests, and spatial hold-out validation. FiLM-U-Net achieved the best performance among the tested models. It obtained an accuracy of 89.73%, an F1-score of 89.41%, and an AUC of 0.953 on the independent test set. In the spatial hold-out validation area, the model achieved an AUC of 0.921. Feature importance analysis showed that distance to roads, rainfall, NDVI, and terrain factors provided important predictive information. These results suggest that the proposed image–tabular joint framework can improve condition-aware feature learning and support regional landslide susceptibility assessment.