This paper proposed an approach, which is based on multi-scale wavelet transform and Gaussian mixture model, to solve the problem about automated fabric defect detection and improve the quality of fabric in the production. Firstly, the sample image was tackled by the “Pyramid” wavelet decomposition algorithm, and the new images were obtained by reconstructing with the produced wavelet coefficients using wavelet thresholding denoising method. Secondly, the obtained new images were segmented by applying the Gaussian mixture model that was based on the Expectation–Maximization (EM) algorithm. Various fabric samples were used in the evaluation, and the experimental results showed that the designed algorithm could precisely locate the position of defect and segment the defect.

This paper aims at investigating methods for solving the problem of automated fabric defect detection and classification, which are more essential and important in assuring the fabric quality. The work focuses on two aspects: fabric defect detection and classification. In the experiment, first, the detection texture features for texture defect are extracted using Gabor filters. The method would automatically segment defects from the regular texture. Second, texture features for classification use local binary patterns and Tamura method. Fabric samples are used in evaluation and the experimental results obtained further confirm the designed algorithm achieved a high detection success rate.

In this paper, we present a robust and reliable framework based on L0 gradient minimization (LGM) and the fuzzy c-means (FCM) method to detect various fabric defects with diverse textures. In our framework, the L0 gradient minimization is applied to process the fabric images to eliminate the influence of background texture and preserve sharpened significant edges on fabric defects. Then, the processed fabric images are clustered by using the fuzzy c-means. Through continuous iterative calculation, the clustering centers of fabric defects and non-defects are updated to realize the defect regions segmentation. We evaluate the proposed method on various samples, which include plain fabric, twill fabric, star-patterned fabric, dot-patterned fabric, box-patterned fabric, striped fabric and statistical-texture fabric with different defect types and shapes. Experimental results demonstrate that the proposed method has a good detection performance compared with other state-of-the-art methods in terms of both subjective and objective tests. In addition, the proposed method is applicable to industrial machine vision detection with limited computational resources.

To evaluate accurately yarn apparent evenness, a simple yet effective method based on modified Canny edge detection was presented. Firstly, the originally captured yarn image is processed by Gaussian filtering, and the weighted least square method is used to calculate the horizontal and vertical directions of the image gradient for obtaining gradient images. Then the non-maximum suppression is applied to process the obtained gradient images and the double threshold algorithm is used to get the yarn edge curve. Finally, the average yarn diameter and the yarn coefficient variation (CV) of yarn were calculated via the number of pixels at the upper and lower edge points of the yarn evenness. To verify the validity and accuracy of the method, yarns with different linear densities were tested. The experimental results demonstrate that the proposed method can accurately detect the unevenness of the yarn apparent diameter.

Sketch-drawing is one of the simplest and most direct means to illustrate 3-D objects. It can not only present geometric features, but also greatly facilitate us to identify and understand the object. This paper presents a simple yet effective tool to generate view-dependent sketch-drawing from point clouds. First, we extract the ridge-valley lines in geometric regions by utilizing their curvature; meanwhile, we extract the contours of point clouds based on the object view-dependent; the obtained ridge-valley lines and contours are optimized and fused to describe the geometric features of the object. Second, we apply the change models of line thickness diffusion based on angle and depth to finish the line thickness variation. Finally, the shadow region of sketch-drawing is achieved by using the shadow generation model based on line density and gray value. Experimental results show that our proposed method can effectively complete the object sketch-drawing simulation from point clouds. Furthermore, our method is more flexible and robust than the existing algorithms; it does not require the preprocessing on the input point clouds.

Deep-learning models have been successfully applied to fabric defect detection fields. However, due to the lack of available dataset for the digital printing fabric defect detection field, the study is still under-explored. Besides, there are two tough challenges that currently existing deep-learning models cannot resolve well. On the one hand, the actual production requires the models to improve its detection accuracy as much as possible while ensuring real-time detection. On the other hand, the objectives of classification and localization are correlated, so the link between classification and localization should be established for an excellent object detection model. To solve these problems, such Classification-Aware Regression Loss (CARL) is embedded into the YOLOF to correlate the classification and localization tasks so that the model is named CARL-YOLOF. The performance of the proposed model has been trained and evaluated by the self-built digital printing fabric defect detection dataset (DPFD-DET). Experimental results demonstrate that CARL-YOLOF achieves 0.54 AP on COCO metrics, which achieves a gain of 0.04 compared with YOLOF. Besides, CARL-YOLOF maintains the speed advantage of YOLOF, which reaches 42 FPS. Therefore, the proposed model is suitable for the real-time digital printing fabric defect detection.