# A Yolo-Based Semantic Segmentation Model for Solar Photovoltaic Panel Identification

## TL;DR

This paper introduces a YOLO-based model to identify solar panels in cities, helping estimate their energy potential with high accuracy.

## Contribution

A novel YOLO-based semantic segmentation framework for city-scale solar panel detection and energy estimation.

## Key findings

- The model achieved 98.32% accuracy in detecting solar panels in complex urban settings.
- The total solar panel area in the Elephant and Castle area of London was estimated at 127.75 m².

## Abstract

The global shift towards renewable energy is increasingly driven by the need to reduce carbon emissions and address urban energy demands sustainably. Solar power, as an accessible and efficient energy source, offers substantial potential for integration within urban environments. However, there remains a lack of a comprehensive evaluation framework for accurately predicting the energy generation of urban solar panel installations. Therefore, in this study, we develop a YOLO-based semantic segmentation framework to estimate the energy generation potential of existing solar panels in a city-scale fashion and use the Elephant andCastle area of London city as a case study. The results demonstrate that the proposed model can detect and segment solar panels in complex urban environments with an accuracy of 98.32%, and the total area of solar panels in the designated area is 127.75 m2.