Real-time implementation of fabric defect detection based on variational automatic encoder with structure similarity
- PDF / 3,398,457 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 72 Downloads / 254 Views
ORIGINAL RESEARCH PAPER
Real‑time implementation of fabric defect detection based on variational automatic encoder with structure similarity Wei Wei1 · Dexiang Deng1 · Lin Zeng1 · Chen Zhang2 Received: 24 October 2019 / Accepted: 18 September 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Automatic detection of fabric defects based on machine vision is an important topic in the quality control of cotton textile factories. There are many kinds of defects in fabric production, it is very difficult to classify the defects automatically. In recent years, deep learning image processing technology based on a convolutional neural network (CNN) can train and extract features of the target image automatically. Since a large number of defect samples cannot be collected completely, we compared unsupervised learning algorithms based on CNN, including auto encoder (AE), variational automatic encoder (VAE), and generative adversarial networks (GAN). Because of the large amount of calculation and the difficulty of training in GAN, we chose AE and VAE codec networks and then introduced mean structural similarity (MSSIM) as network training loss function to improve the performance that only used Lp-distance loss function for image brightness comparison. After training finished, the authors used the trained model to obtain target defects from SSIM residual maps between input and reconstruct images. According to the evaluation results, we finally implemented a fabric defect detection system based on VAE on Jetson TX2 from Nvidia Corporation, USA. The optimized algorithm can meet the real-time requirements of the project and realize its popularization and application. Keywords Fabric defects · Variational automatic encoder · Structural similarity · Real-time · Deep learning
1 Introduction The detection of fabric defect is an important procedure in modern textile enterprises, the number of fabrics with defects is the main factor affecting the quality of fabrics, and therefore, the effective detection of fabric defect is extremely important. With the development of computer technology, machine vision has been widely used in industrial inspection [1]. With the definition of machine vision, from describing the texture features of fabrics, there are four kinds of algorithms: statistical method, frequency-domain method, model method, and machine learning method, but there are always several kinds of methods involved in practical application. Statistical methods describe the overall texture * Wei Wei [email protected] 1
School of Electronic Information, Wuhan University, Wuhan 430072, China
School of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan 430072, China
2
characteristics of fabric by analyzing texture features of statistical target samples, such as the spatial distribution of gray values of pixels, quantitative indicators, etc. The frequency-domain method is a classical algorithm in fabric detection [2–4]. It has a good effect to use a multi-resolution anal
Data Loading...
