SIGAN: A Novel Image Generation Method for Solar Cell Defect Segmentation and Augmentation

Abstract

Solar cell electroluminescence (EL) defect segmentation is an interesting and challenging topic. Many methods have been proposed for EL defect detection, but these methods are still unsatisfactory due to the diversity of the defect and background. In this paper, we provide a new idea of using generative adversarial network (GAN) for defect segmentation. Firstly, the GAN-based method removes the defect region in the input defective image to get a defect-free image, while keeping the background almost unchanged. Then, the subtracted image is obtained by making difference between the defective input image with the generated defect-free image. Finally, the defect region can be segmented through thresholding the subtracted image. To keep the background unchanged before and after image generation, we propose a novel strong identity GAN (SIGAN), which adopts a novel strong identity loss to constraint the background consistency. The SIGAN can be used not only for defect segmentation, but also small-samples defective dataset augmentation. Moreover, we release a new solar cell EL image dataset named as EL-2019, which includes three types of images: crack, finger interruption and defect-free. Experiments on EL-2019 dataset show that the proposed method achieves 90.34% Fscore, which outperforms many state-of-the-art methods in terms of solar cell defects segmentation results.

A high-effective multitask surface defect detection method based on CBAM and atrous convolution

“…Then, the subtracted image is obtained by taking the difference between the defective input image and the generated defect-free image. Finally, the defect region can be segmented by thresholding the subtracted image (Su et al, 2021). Lin et al proposed Normal Background Regularization and Crop-and-Paste operations, using abundant Normal image training models without defects to segment defect data with few-shot learning, to prevent model overfitting caused by very limited defect training images (Lin et al, 2020).…”