Electroluminescence (EL) imaging is a widely adopted method in quality assurance of the photovoltaic (PV) manufacturing industry. With the growing demand for high-quality PV products, automatic inspection methods based on machine vision have become an emerging area concern to replace manual inspectors. Therefore, this paper presents an automatic defect-inspection method for multi-cell monocrystalline PV modules with EL images. A processing routine is designed to extract the defect features of the PV module, eliminating the influence of the intrinsic structural features. Spectrum domain analysis is applied to effectively reconstruct an improved PV layout from a defective one by spectrum filtering in a certain direction. The reconstructed image is used to segment the PV module into cells and slices. Based on the segmentation, defect detection is carried out on individual cells or slices to detect cracks, breaks, and speckles. Robust performance has been achieved from experiments on many samples with varying illumination conditions and defect shapes/sizes, which shows the proposed method can efficiently distinguish intrinsic structural features from the defect features, enabling precise and speedy defect detections on multi-cell PV modules.

Nanoscale coating manufacturing (NCM) process modeling is an important way to monitor and modulate coating quality. The multivariable prediction of coated film and the data augmentation of the NCM process are two common issues in smart factories. However, there has not been an artificial intelligence model to solve these two problems simultaneously. Focusing on the two problems, a novel auxiliary regression using a self-attention-augmented generative adversarial network (AR-SAGAN) is proposed in this paper. This model deals with the problem of NCM process modeling with three steps. First, the AR-SAGAN structure was established and composed of a generator, feature extractor, discriminator, and regressor. Second, the nanoscale coating quality was estimated by putting online control parameters into the feature extractor and regressor. Third, the control parameters in the recipes were generated using preset parameters and target quality. Finally, the proposed method was verified by the experiments of a solar cell antireflection coating dataset, the results of which showed that our method performs excellently for both multivariable quality prediction and data augmentation. The mean squared error of the predicted thickness was about 1.6~2.1 nm, which is lower than other traditional methods.

Self-taught learning (STL) is a promising solution that reduces the performance gap between weakly supervised and fully supervised learning for easily accessible, label-free images. The success of traditional STL solutions relies on the assumption that the target appearance is completely visible and well-defined. In real-world facial expression recognition scenarios, however, saliency regions are often partially occluded, which significantly hampers the generalization capability of STL methods. Nevertheless, few studies have investigated the impact of occlusion on STL. In this paper, we propose an interweaved autoencoder network for weakly supervised facial expression recognition in occlusion scenarios. The key innovation of our network lies in the Residual Connection Union (RCU) blocks that can integrate the Convolutional Neural Network (CNN) and Transformer layers into a multi-scale structure. The RCU enables a progressive masking strategy to accurately identify and focus on contributive yet often overlooked image patches by analyzing the relationships among region-level target representations. In addition, we introduce a self-knowledge distillation module for the effective training of the proposed autoencoder network. Extensive experiments are conducted on four public datasets to demonstrate the superiority of our method over related works.