Home/MEJ - Mansoura Engineering Journal/Article Record
Peer-Reviewed Article February 28, 2025

Applying Image Processing and Computer Vision for Damage Detection in Photovoltaic Panels

Are you an author of this work?Link to your sovereign AfriScholar profile
Published in: MEJ - Mansoura Engineering Journal
Publisher: Mansoura University, Faculty of Engineering • Egypt • 1110-0923 • 2735-4202
DOI: https://doi.org/10.58491/2735-4202.3263

Abstract

As the use of Photovoltaic (PV) panels as a sustainable energy source grows, there is a need for effective and precise techniques to monitor and manage these systems. Conventional techniques for identifying faults in PV panels, such as manual inspections, require a significant amount of effort and are susceptible to mistakes made by humans. This study introduces an innovative automated method that utilizes image processing techniques implemented using the OpenCV library to identify panel faults, namely hotspots, which are important indications of possible failures. The study approach combines grayscale conversion, histogram analysis, and adaptive thresholding to accurately detect and evaluate abnormalities in panel integrity. The method starts with the conversion of Red, Blue, and Green (RGB) images to grayscale in order to decrease computing complexity. Histogram analysis then follows to evaluate the pixel intensity distribution. The study then employs adaptive thresholding to accurately delineate affected regions, thereby improving the overall precision of damage identification. By reducing time and expense, this automated technology enhances both the precision of detection and the efficiency of existing approaches. The results indicate that automated image processing is a scalable and efficient technique for regular maintenance of PV panels, which has the ability to avert significant damage and optimize panel performance. Future research endeavors will involve the integration of sophisticated Deep Learning (DL) algorithms such as ResNet and YOLO to improve system detection capabilities and automation, thereby facilitating the development of more robust energy infrastructure

Download Full Text PDFView at PublisherCrossref RecordCrossref XML

Verified Citation Velocity

Recorded scholarly citations in African and international peer-reviewed literature

26
Total Citations

Citations are continuously tracked and harvested across continental OJS endpoints and OpenAlex feeds.