Solar panels are crucial for solar power systems. Their quality affects how well they generate power and how safe the system is. Current ways to find defects in solar panels have issues like high costs, heavy workloads, and slow detection speeds. To fix these problems, scientists have created a new method called ST-YOLO. It uses infrared thermal imaging and machine vision technology to spot defects quickly and accurately.

ST-YOLO is based on YOLOv8s, a popular detection algorithm. It introduces a special part called C2f-SCconv convolution module. This module is designed to be lightweight, which means it requires less computing power. This makes the detection process faster. Additionally, ST-YOLO includes a Triplet Attention mechanism. This mechanism improves the accuracy of defect detection without making the model too heavy.

Tests were done on a set of infrared images of solar panels. ST-YOLO performed better than other methods like YOLOv8s, YOLOv7-Tiny, and YOLOv5s. It had fewer model weights, meaning it needed less computing power. It also showed better precision and mAP@0. 5 scores. mAP@0. 5 is a measure of how well the model can detect objects.

These results show that ST-YOLO is a big step forward in detecting defects in solar panels. It makes the process faster and more accurate, which is good news for the solar power industry.

Catching Solar Panel Flaws: A New Way with ST-YOLO

Actions