Deep learning improves 3D mapping of power lines from aerial scans
Learning-Based 3D Reconstruction of Power Networks from Aerial Point Clouds
Computer Vision and Pattern RecognitionMachine Learning
Summary
Mapping overhead power lines and poles from aerial 3D scans is hard because wires are thin and scenes are cluttered. The authors created a method that first identifies poles and wires using a special neural network, then figures out which poles are connected by wires using a combination of geometric rules and a small deep learning model. This approach is better at correctly recognizing connections, especially in tricky areas. It also estimates useful details like wire height and sag, helping with power network management.
What this means in practice
- •For utility maintenance teams: Automatically map power network components and their connections from aerial LiDAR to support inspection and repair planning.
- •For civil engineering firms: Generate detailed 3D models of overhead power lines and poles for infrastructure design and evaluation from large-scale aerial data.
Authors
Rishabh Jain, Anuja Saini, Vishal Jain
Abstract
This paper presents an end-to-end framework for reconstructing overhead power utility network topology and extracting span-level physical metadata from large-scale aerial LiDAR. The pipeline begins with semantic segmentation of the input point cloud using an improved KPConv-based model, in which data sampling and loss functions are adapted to emphasize pole and conductor (wire) classes. Network topology inference then proceeds in two stages: (i) pole instances are obtained by clustering pole-class points and validating candidates using geometric criteria, including height and verticality estimated via PCA, and (ii) candidate pole pairs are evaluated using a heuristic method and a lightweight ResNet-based classifier on 2D top-view projections of pole and wire point distributions to determine whether a physical conductor span exists. By explicitly classifying candidate spans, the approach mitigates common failure modes of heuristic connectivity rules in dense or cluttered scenes and under partial wire observation. For each validated wire, attributes regarding utility infrastructure geometry are computed, including endpoint conductor heights, ground elevation, sag-related lowest-point features, conductor arrangement, and wire width. Evaluation on multiple real-world aerial LiDAR datasets demonstrates decimeter-level endpoint height accuracy and approximately 9% relative improvement in recall for topology reconstruction compared to heuristic nearest-neighbor baselines, with larger gains in complex layouts.