This project focuses on developing a deep learning-based system for detecting defects in solar panels using electroluminescence images. The system employs advanced object detection algorithms, YOLOv8 and YOLOv9, to identify and classify multiple types of defects including black core, corner crack, finger fragment, horizontal dislocation, printing error, scratch, short circuit, star crack, thick line, and vertical dislocation. The proposed solution integrates a web-based interface built with Flask, allowing users to upload images and perform live detection. The system ensures accurate labeling and visualization of defects, providing a structured approach to panel analysis. The modules include registration, login, image upload, live detection, and logout. By combining deep learning and interactive interface design, the system provides an automated, precise, and efficient method for analyzing electroluminescence images. This project contributes to defect identification research and facilitates structured inspection processes.

Keywords: deep learning, electroluminescence, solar panel, defect detection, YOLOv8, YOLOv9, Flask, image analysis, object detection, automated inspection