Semi-supervised defect classification of steel surface based on multi-training and generative adversarial network

•A semi-supervised learning (SSL) defect classification method is proposed to inspect defect on the steel surface.•The proposed SSL method can generate a large number of unlabeled samples that assist the labeled ones in learning.•The multi-training algorithm of two deep networks is proposed to include the unlabeled samples into semi-supervised learning.•The proposed SSL method can achieve promising results even if the original defect samples are quite few.•This work provide a new way to apply large and strong deep learning networks in data-limited industrial inspection tasks. Defect inspection is very important for guaranteeing the surface quality of industrial steel products, but related methods are based primarily on supervised learning which requires ample labeled samples for training. However, there can be no doubt that inspecting defects on steel surface is always a data-limited task due to difficult sample collection and expensive expert labeling. Unlike the previous works in which only labeled samples are treated using supervised classifiers, we propose a semi-supervised learning (SSL) defect classification approach based on multi-training of two different networks: a categorized generative adversarial network (GAN) and a residual network. This method uses the GAN to generate a large number of unlabeled samples. And then the multi-training algorithm that uses two classifiers based on different learning strategies is proposed to integrate both labeled and unlabeled into SSL process. Finally, through the multiple training process, our SSL method can acquire higher accuracy and better robustness than the supervised one using only limited labeled samples. Experimental results clearly demonstrate that the effectiveness of our proposed method, achieving the classification accuracy of 99.56%.