Old landslides in the Loess Plateau, Northwest China usually occurred over a relatively long period, and their sizes are usually smaller compared to old landslides in the alpine valley areas of Sichuan, Yunnan, and Southeast Tibet. These landslide areas may have been changed either partially or greatly, and they are usually covered with vegetation and similar to their surrounding environment. Therefore, it is a great challenge to detect them using high-resolution remote sensing images with only orthophoto view. This paper proposes the optimal-view and multi-view strategic hybrid deep learning (OMV-HDL) method for old loess landslide detection. First, the optimal-view dataset in the Yan’an area (YA-OP) was established to solve the problem of insufficient optical features in orthophoto images. Second, in order to make the process of interpretation more labor-saving, the optimal-view and multi-view (OMV) strategy was proposed. Third, hybrid deep learning with weighted boxes fusion (HDL-WBF) was proposed to detect old loess landslides effectively. The experimental results with the constructed optimal-view dataset and multi-view data show that the proposed method has excellent performance among the compared methods—the F1 score and AP (mean) of the proposed method were improved by about 30% compared with the single detection model using traditional orthophoto-view data—and that it has good detection performance on multi-view data with the recall of 81.4%.

In high-mountain canyon areas, complex geological environments lead to frequent cascading disasters with unclear triggering mechanisms, posing severe threats to human life and property. Existing knowledge graph research in geology predominantly focuses on single-hazard types or general geological entities, lacking structured modeling and specialized datasets for cascading disaster processes, particularly the evolutionary chains in high-mountain canyon settings. To address this gap, this study proposes a method for constructing a knowledge graph tailored to cascading disasters in high-mountain canyon regions. First, a three-layer schema framework—comprising concept, relation, and instance layers—was designed to systematically characterize the knowledge elements and evolutionary relationships of disaster chains. To address the lack of a knowledge dataset for cascade disasters, this paper integrates multi-source heterogeneous data to construct a high-mountain canyon cascading disasters entity–relation dataset (DCER-MC), providing a reliable benchmark for related tasks. Based on this dataset, we implemented the knowledge graph and conducted disaster chain analysis. Experiments and applications demonstrate that the constructed knowledge graph effectively supports structured storage, centralized management, and scenario-based application of regional cascading disaster information. The main contributions of this work are (1) proposing a targeted schema framework for cascading-disaster knowledge graphs; (2) releasing a specialized dataset for cascading disasters in high-mountain canyon regions; and (3) establishing a complete pipeline from data to knowledge to scenario-based services, offering a novel knowledge-driven paradigm for disaster chain risk identification, inference prediction, and emergency decision-making in these areas.