期刊
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
卷 72, 期 -, 页码 -出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIM.2022.3232649
关键词
Data augmentation; defect image generation; generative adversarial network (GAN); limited data; surface defect recognition
Surface defect recognition is crucial in intelligent manufacturing, and deep learning is a popular method for this task. However, the lack of available defective samples poses challenges for deep learning methods, so generative adversarial networks are used to generate synthetic samples. To improve training and image quality, a new GAN called contrastive GAN is proposed, which generates diverse defects with limited samples. Experimental results show that the proposed GAN generates higher quality defective images and improves the accuracy of DL networks.
Surface defect recognition (SDC) is essential in intelligent manufacturing. Deep learning (DL) is a research hotspot in SDC. Limited defective samples are available in most real-world cases, which poses challenges for DL methods. Given such circumstances, generating defective samples by generative adversarial networks (GANs) is applied. However, insufficient samples and high-frequency texture details in defects make GANs very hard to train, yield mode collapse, and poor image quality, which can further impact SDC. To solve these problems, this article proposes a new GAN called contrastive GAN, which can be trained to generate diverse defects with only extremely limited samples. Specifically, a shared data augmentation (SDA) module is proposed for avoiding overfitting. Then, a feature attention matching (FAM) module is proposed to align features for improving the quality of generated images. Finally, a contrastive loss based on hypersphere is employed to constrain GANs to generate images that differ from the traditional transform. Experiments show that the proposed GAN generates defective images with higher quality and lower variance between real defects compared to other GANs. Synthetic images contribute to pretrained DL networks with accuracies of up to 95.00%-99.56% for Northeastern University (NEU) datasets of different sizes and 91.84% for printed circuit board (PCB) cases, which proves the effectiveness of the proposed method.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据