Solar Panel Mapping via Oriented Object Detection

TL;DR

This paper presents a deep learning method for automatically detecting and mapping individual solar panels in power plants, significantly improving efficiency and scalability for maintaining solar infrastructure.

Contribution

The paper introduces an end-to-end rotated object detection framework specifically designed for solar panel mapping, achieving high accuracy on diverse datasets.

Findings

Achieved a mean Average Precision (mAP) of 83.3% on a diverse US dataset.

Demonstrated the effectiveness of rotated object detection for solar panel identification.

Provided a scalable solution for solar plant maintenance and monitoring.

Abstract

Maintaining the integrity of solar power plants is a vital component in dealing with the current climate crisis. This process begins with analysts creating a detailed map of a plant with the coordinates of every solar panel, making it possible to quickly locate and mitigate potential faulty solar panels. However, this task is extremely tedious and is not scalable for the ever increasing capacity of solar power across the globe. Therefore, we propose an end-to-end deep learning framework for detecting individual solar panels using a rotated object detection architecture. We evaluate our approach on a diverse dataset of solar power plants collected from across the United States and report a mAP score of 83.3%.