Tempo di lettura ca.: 1 minuti, 26 secondi
[[{“value”:”
The novel technique is based on the VarifocalNet deep-learning object detection framework, which was reportedly tweaked to achieve quicker and more accurate results. Compared to other such methods, the new approach was found to be the most accurate and third quickest.
A research group from China’s Beihua University and the Northeast Electric Power University has developed a novel PV defect detection method based on deep learning of electroluminescence (EL) imaging.
“Defects in PV cells can lead to module failure, which can result in reduced power output and pose safety risks to the system,” said the researchers. “Therefore, it is essential to conduct regular inspections and maintenance of photovoltaic modules to ensure maximum output from the PV system throughout its lifespan.”
The proposed method is based on the VarifocalNet deep-learning object detection framework, which is a method aimed at accurately ranking a huge number of candidate detections in object detection. Deep convolutional neural network ResNet-101 is used as the backbone for feature extraction. To make the VarifocalNet faster, the group designed a bottleneck module with smaller parameters, that is, a layer of the neural network designed to reduce the number of parameters and computational complexity.
“To improve detection accuracy, we design a bottleneck module without reducing feature map size to replace the first bottleneck module used in the last stage convolution group of backbone in VarifocalNet,” the academics explained. “To further improve detection accuracy, we also designed a new feature interactor and improved the regression loss function.”
The new detection method was trained and tested on the PVEL-AD dataset, which contains 4,0000 near-infrared images featuring a range of defects, including cracks, finger scratches, black cores, and horizontal dislocations. In addition, for reference, other detection methods were tested on the same database: RetinaNet, DDH-YOLOv5, Faster GG R-CNN, Cascade R-CNN, the unimproved VarifocalNet, Improved Faster R-CNN, and improved YOLOv7.
In EL images, defects appear as dark gray lines and areas, and the novel method is said to detect them quickly and accurately.
“Our method exhibits the highest mean average precision (mAP) and Recall, indicating that the defect detection accuracy of our method is higher than that of other methods,” the team stated. “Additionally, it also has a faster detection speed than other methods except for the DDH-YOLOv5 method and the improved YOLOv7 method.”
The researchers highlighted that while their model is a two-stage method, both DDH-YOLOv5 and the improved YOLOv7 belong to the one-stage method. “The two-stage method features a complex network structure, resulting in higher detection accuracy and slower detection speed, whereas the one-stage method employs a relatively simpler network structure, leading to faster detection speed and lower detection accuracy,” they further explained.
The novel approach was described in the paper “Defect detection of photovoltaic modules based on improved VarifocalNet,” published in Scientific Reports.
“}]]