Advances in Remote Sensing and Artificial Intelligence for Vegetation Monitoring in Power Line Corridors: A review and future directions

Remote sensing techniques, coupled with intelligent data analytics, have become invaluable for monitoring power lines and mitigating risks posed by vegetation. These technologies transform sensed data into actionable insights for decision makers, though the landscape of methods remains fragmented. This article provides a comprehensive and structured review of remote sensing methods integrated with artificial intelligence (AI) for identifying vegetation hazards near power lines. It emphasizes the role of 2D and lidar-based data combined with AI in detecting power lines and nearby vegetation, explores image analysis for extracting useful information, and consolidates key performance metrics and publicly available datasets. While lidar combined with machine learning (ML) has demonstrated significant potential, the limited availability of public datasets has hindered the development of robust 3D deep learning (DL) models. In this context, the article also outlines future directions, including lidar-specific data analysis, 2D/3D data fusion, learning from limited datasets, and edge computing for real-time applications. These advancements offer significant potential for enhancing the understanding of the challenges and opportunities in processing and analyzing remote sensing data for vegetation monitoring in power line corridors.