Solar cells defects inspection plays an important role to ensure the efficiency and lifespan of photovoltaic modules. However, it is still an arduous task because of the diverse attributes of electroluminescence images, such as indiscriminative complex background with extremely unbalanced defects and various types of defects. In order to deal with these problems, this paper proposes a new precise and accurate defect inspection method for photovoltaic electroluminescence (EL) images. The proposed algorithm leverages the advantage of multi attention network to efficiently extract the most important features and neglect the nonessential features during training. Firstly, we designed a channel attention to exploit contextual representations and spatial attention to effectively suppress background noise. Secondly, we incorporate both attention networks into modified U-net architecture and named it multi attention U-net (MAU-net) to extract effective multiscale features for defects inspection. Finally, we propose a hybrid loss which combines focal loss and dice loss aiming to solve two problems: a) overcome the class imbalance problem, and b) allowing the network to train with irregular image labels for some complex defects. The proposed multi attention U-net is evaluated on real photovoltaic EL images datasets using 5-fold cross validation technique. Experimental results demonstrate that the proposed network can segment and detect various complex defects correctly. The proposed method achieved the mean intersection over-union (m-IOU) of 0.699 and F-measure of 0.799 which outperforms the previous methods.

Automatic defect detection of solar cells' near-infrared electroluminescence (EL) images is a challenging task due to non-uniform complex texture background interference and the defects' small size. Therefore, this paper proposes a new surface defect detection scheme of solar cells based on Faster-RCNN. In order to improve the accuracy and speed of defect detection, we modify the defect extraction part and RPN part of the original Faster-RCNN. Firstly, we combine CNN with Feature Pyramid Network (FPN) to achieve multi-scale defect feature extraction. By combining multiple layers of feature information, the semantic information is more robust and the detection accuracy of small targets is improved. Secondly, we apply Guided Anchoring RPN (GA-RPN) to achieve adaptive adjustment of the anchor. It will predict the position and shape of the anchor according to the feature of the defect, and then use the feature adaption to adjust the feature map to generate proposals with high quality and low density. Experiment results show that the improved model has a mAP of 94.62, an increase of 11.26% compared with the original Faster-RCNN model, and the detection speed is increased from 0.91s/img to 0.19s/img, which can meet the basic requirements of industrial production. The architecture combining FPN, GA-RPN and Faster-RCNN improves the accuracy and speed of defect detection, and effectively realizes the defect detection of polysilicon near-infrared image under complex background, which lays a foundation for the defect detection automation of solar cells.

High-quality and fast crack segmentation in solar cell electroluminescence images with heterogeneously textured background disturbance is challenging. We propose an end-to-end Efficient and Refined Deep Convolutional Features Network (ERDCF-Net) for precise and efficient crack segmentation. Firstly, we design a lightweight, efficient deep convolutional features (EDCF) encoder by adopting multilevel deep convolutional features to generate a discriminative feature representation of cracks. Moreover, some complicated background disturbance is restrained by reusing rich crack and background features. Secondly, we present a refined side output (RS) decoder by automatically refining side outputs from the EDCF encoder. The experiments demonstrate that the proposed network excellently performs on the solar cell cracks (MIoU of 92.82%, F -measure of 93.58%, and 89 FPS) and open crack datasets (MIoU of 85.6%, F -measure of 84.1%, and 90 FPS).

The task of collaborative human pose forecasting stands for predicting the future poses of multiple interacting people, given those in previous frames. Predicting two people in interaction, instead of each separately, promises better performance, due to their body-body motion correlations. But the task has remained so far primarily unexplored.In this paper, we review the progress in human pose forecasting and provide an in-depth assessment of the single-person practices that perform best for 2-body collaborative motion forecasting. Our study confirms the positive impact of frequency input representations, space-time separable and fully-learnable interaction adjacencies for the encoding GCN and FC decoding. Other single-person practices do not transfer to 2-body, so the proposed best ones do not include hierarchical body modeling or attention-based interaction encoding.We further contribute a novel initialization procedure for the 2-body spatial interaction parameters of the encoder, which benefits performance and stability. Altogether, our proposed 2-body pose forecasting best practices yield a performance improvement of 21.9% over the state-of-the-art on the most recent ExPI dataset, whereby the novel initialization accounts for 3.5%. See our project page at https://www.pinlab.org/bestpractices2body