Fusing Multilevel Deep Features for Fabric Defect Detection Based NTV-RPCA

Abstract

Fabric defect detection plays an important role in automated inspection and quality control in textile manufacturing. As the fabric images have complex and diverse textures and defects, traditional detection methods show a poor adaptability and low detection accuracy. Robust principal component analysis (RPCA) model that can be used to separate the image into object and background have proven applicable in fabric defect detection. However, how to represent texture feature of the fabric image more effectively is still problematic in this kind of method. In addition, the use of the traditional RPCA may result in low accuracy and more noises in sparse part. In this paper, a novel fabric defect detection method based on multilevel deep features fusion and non-convex total variation regularized RPCA (NTV-RPCA) is proposed. Firstly, the image representation ability is well enhanced through multilevel deep features extracted by a convolutional neural network. Then, the non-convex total variation regularized RPCA is proposed in which total variation constraint significantly reduces the noises in sparse part and non-convex solution is more approximate to the authentic one. Next, multilevel saliency maps generated by the sparse matrixes are fused via RPCA to produce a more reliable detection result. Finally, the defect region is located by segmenting the fused saliency map via a threshold segmentation algorithm. Qualitative and quantitative experiments conducted on two public fabric image databases demonstrate that the proposed method improves the adaptability and detection accuracy comparing to the state-of-the-arts.

Deep learning-based fabric defect detection: A review

“…Dong et al. 61 proposed a multilevel fully convolutional feature and nonconvex total variation regularized robust principal component analysis (RPCA)-based fabric defect detection approach. As an initial step of this study, VGG16 improves the image representation significantly.…”

“…Investigated papers were analyzed and summarized in terms of method, dataset, classification or number of classes, performance as success, and comparison. 15,24,44–49,52–77…”

MACN: A cascade defect detection for complex background based on mixture attention mechanism

“…[4] and Dong et al. [5] used a non‐convex total variational regularized nucleated robust principal component analysis (RPCA) to better deal with the messy and complex background in low‐rank space, based on the characteristics of an RPCA model that can separate images into target and background. Classifiers can be used in addition to the methods mentioned before.…”

“…However, in the presence of some complicated patterns, such as cloth and rail defects, it is still difficult to distinguish the defective area from the background. Wang et al [4] and Dong et al [5] used a non-convex total variational regularized nucleated robust principal component analysis (RPCA) to better deal with the messy and complex background in lowrank space, based on the characteristics of an RPCA model that can separate images into target and background. Classifiers can be used in addition to the methods mentioned before.…”

CNN-Based Fabric Defect Detection System on Loom Fabric Inspection

“…Especially by replacing the last part of the mesh with the Deformable Convolution (DC) block structure, the classification success for fabric defects has been increased. In another study, effective results were obtained on two different fabric datasets by using principal component analysis and deep feature extraction strategy together [29]. First of all, the features of the images were extracted with the VGG16 deep learning model.…”