Intelligent Environmental Hazards Detection in Transmission Line Construction Area Based on Remote Sensing and GIS

Traditional manual environmental hazard detection in transmission line construction areas suffers from inefficiency and low accuracy, while intelligent methods often lack positioning precision. To address this, we propose an intelligent detection method integrating remote sensing and GIS technologies. Synthetic aperture radar interferometry (SAR) collects high‐resolution images of construction zones. Processing this SAR data with Small Baseline Subset InSAR (SBAS‐InSAR) generates differential interferograms revealing terrain displacement. Stacking technology enhances these images, mitigating spatiotemporal incoherence and atmospheric delays, improving data quality and reliability. Key environmental hazard indicators (e.g., vegetation coverage, pollution areas) are extracted from the processed imagery. These indicators are input into a specifically designed Asymmetric Convolutional Neural Network (ACNN) model for hazard identification and classification. Experiments validate that the method accurately detects transmission line distribution and effectively assesses the degree of environmental hazards. This integrated approach—combining remote sensing, GIS, advanced image processing, and deep learning—provides a comprehensive and efficient solution for environmental hazard detection in transmission line construction zones, offering insights for mitigation and management.