Global climate change has led to a marked increase in the frequency of record-breaking extreme rainfall events, which often surpass historical benchmarks and pose significant challenges to conventional geological hazard risk assessment methods. This study used a record-breaking extreme rainfall event in Zhenba County, Shaanxi Province, in July 2023 as a case study to develop a tailored risk assessment framework for geological hazards under extreme rainfall conditions. By integrating high-resolution Planet satellite imagery, millimeter-scale surface deformation data derived from SBAS-InSAR, and detailed field investigation results, a comprehensive disaster inventory containing 1012 landslides was compiled. The proposed framework integrates cumulative extreme rainfall metrics with subtle ground deformation indicators and applies four advanced machine learning algorithms—DNN, XGBoost, RF, and LightGBM—for multidimensional hazard assessment. Among these, the DNN model exhibited the highest performance, achieving an AUC of 0.82 and Kappa coefficients of 0.833 (training) and 0.812 (prediction). Further analysis using SHAP values identified distance to rivers, cumulative rainfall, and the Topographic Wetness Index (TWI) as the most influential factors governing landslide occurrence under extreme rainfall conditions. Validation using representative case studies confirmed that the framework effectively identifies high-hazard zones, particularly in areas severely impacted by debris flows and landslide deformation zones. These findings provide a robust scientific foundation and technical basis for early warning, disaster prevention, and mitigation strategies in geologically complex regions increasingly affected by extreme rainfall events.

The rapid advancement of high-resolution remote sensing technology has significantly expanded observational capabilities in the oil and gas sector, enabling more precise identification of petroleum infrastructure. Remote sensing now plays a critical role in providing real-time, continuous monitoring. Manual interpretation remains the predominant approach, yet is plagued by multiple limitations. To overcome the limitations of manual interpretation in large-scale monitoring of upstream petroleum assets, this study develops an end-to-end, deep learning-driven framework for intelligent extraction of key oilfield targets from high-resolution remote sensing imagery. Specific aims are as follows: (1) To leverage temporal diversity in imagery to construct a representative training dataset. (2) To automate multi-class detection of well sites, production discharge pools, and storage facilities with high precision. This study proposes an intelligent monitoring framework based on deep learning for the automatic extraction of petroleum-related features from high-resolution remote sensing imagery. Leveraging the temporal richness of multi-temporal satellite data, a geolocation-based sampling strategy was adopted to construct a dedicated petroleum remote sensing dataset. The dataset comprises over 8000 images and more than 30,000 annotated targets across three key classes: well pads, production ponds, and storage facilities. Four state-of-the-art object detection models were evaluated—two-stage frameworks (Faster R-CNN, Mask R-CNN) and single-stage algorithms (YOLOv3, YOLOv4)—with the integration of transfer learning to improve accuracy, generalization, and robustness. Experimental results demonstrate that two-stage detectors significantly outperform their single-stage counterparts in terms of mean Average Precision (mAP). Specifically, the Mask R-CNN model, enhanced through transfer learning, achieved an mAP of 89.2% across all classes, exceeding the best-performing single-stage model (YOLOv4) by 11 percentage points. This performance gap highlights the trade-off between speed and accuracy inherent in single-shot detection models, which prioritize real-time inference at the expense of precision. Additionally, comparative analysis among similar architectures confirmed that newer versions (e.g., YOLOv4 over YOLOv3) and the incorporation of transfer learning consistently yield accuracy improvements of 2–4%, underscoring its effectiveness in remote sensing applications. Three oilfield areas were selected for practical application. The results indicate that the constructed model can automatically extract multiple target categories simultaneously, with average detection accuracies of 84% for well sites and 77% for production ponds. For multi-class targets over 100 square kilometers, manual detection previously required one day but now takes only one hour.