STN PLAD: A Dataset for Multi-Size Power Line Assets Detection in High-Resolution UAV Images

TL;DR

This paper introduces the STN PLAD dataset, a high-resolution collection of annotated power line assets captured via UAVs, to facilitate the development of better detection methods for diverse power line components.

Contribution

The work provides a new publicly available dataset with diverse, annotated high-resolution UAV images of power line assets, enabling advancements in detection algorithms.

Findings

Deep detection methods show room for improvement on this dataset.

The dataset includes 2,409 annotated objects across five classes.

Images vary in size, orientation, and background, presenting real-world challenges.

Abstract

Many power line companies are using UAVs to perform their inspection processes instead of putting their workers at risk by making them climb high voltage power line towers, for instance. A crucial task for the inspection is to detect and classify assets in the power transmission lines. However, public data related to power line assets are scarce, preventing a faster evolution of this area. This work proposes the Power Line Assets Dataset, containing high-resolution and real-world images of multiple high-voltage power line components. It has 2,409 annotated objects divided into five classes: transmission tower, insulator, spacer, tower plate, and Stockbridge damper, which vary in size (resolution), orientation, illumination, angulation, and background. This work also presents an evaluation with popular deep object detection methods, showing considerable room for improvement. The STN PLAD dataset is publicly available at https://github.com/andreluizbvs/PLAD.