Textile defect detection using multilevel and attentional deep learning network (MLMA-Net)

Abstract

For the sake of recognizing and classifying textile defects, deep learning-based methods have been proposed and have achieved remarkable success in single-label textile images. However, detecting multilabel defects in a textile image remains challenging due to the coexistence of multiple defects and small-size defects. To address these challenges, a multilevel, multi-attentional deep learning network was proposed and built to: (a) increase the feature representation ability to detect small-size defects; and (b) generate discriminative representation that maximizes the capability of attending the defect status, which leverages higher-resolution feature maps for multiple defects. Moreover, a multilabel object detection dataset in textile defect images was built to verify the performance of the proposed model. The results demonstrate that the network extracts more distinctive features and has better performance than the state-of-the-art approaches on the real-world industrial dataset.