Synthetic data augmentation for surface defect detection and classification using deep learning
- PDF / 1,913,034 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 88 Downloads / 205 Views
Synthetic data augmentation for surface defect detection and classification using deep learning Saksham Jain1 · Gautam Seth1 · Arpit Paruthi1 · Umang Soni1
· Girish Kumar2
Received: 29 October 2019 / Accepted: 29 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Deep learning techniques, especially Convolutional Neural Networks (CNN), dominate the benchmarks for most computer vision tasks. These state-of-the-art results are typically obtained through supervised learning, for which large annotated datasets are required. However, acquiring such datasets for manufacturing applications remains a challenging proposition due to the time and costs involved in their collection. To overcome this disadvantage, a novel framework is proposed for data augmentation by creating synthetic images using Generative Adversarial Networks (GANs). The generator synthesizes new surface defect images from random noise which is trained over time to get realistic fakes. These synthetic images can be used further for training of classification algorithms. Three GAN architectures are trained, and the entire data augmentation pipeline is implemented for the Northeastern University (China) Classification (NEU-CLS) dataset for hot-rolled steel strips from NEU Surface Defect Database. The classification accuracy of a simple CNN architecture is measured on synthetic augmented data and further it is compared with similar state-of-the-arts. It is observed that the proposed GANs-based augmentation scheme significantly improves the performance of CNN for classification of surface defects. The classically augmented CNN yields sensitivity and specificity of 90.28% and 98.06% respectively. In contrast, the synthetically augmented CNN yields better results, with sensitivity and specificity of 95.33% and 99.16% respectively. Also, the use of GANs is demonstrated to disentangle the representation space and to add additional domain knowledge through synthetic augmentation that can be difficult to replicate through classic augmentation. The proposed framework demonstrates high generalization capability. It may be applied to other supervised surface inspection tasks, and thus facilitate the development of advanced vision-based inspection instruments for manufacturing applications. Keywords Surface defects · Classification · Convolutional neural network · Generative adversarial network · Deep learning
Introduction In today’s world, the industry is facing a challenge in the use of supervised learning algorithms for classification. Despite the various advantages of these learning algorithms, their performance decreases when trained on a real-world dataset. One reason is a shortage of training data due to factors that are typical to the manufacturing industry. With the advancements made in modern manufacturing processes and implementation of quality programs including six sigma, the number of good samples greatly exceeds the number of defective samples. The number of defectives are sometimes to the tune of
B
Umang Soni
Data Loading...
