Henry Y. T. Ngan

Review article: Automated fabric defect detection-A review

This paper provides a review of automated fabric defect detection methods developed in recent years. Fabric defect detection, as a popular topic in automation, is a necessary and essential step of quality control in the textile manufacturing industry. In categorizing these methods broadly, a major group is regarded as non-motif-based while a minor group is treated as motif-based. Non-motif-based approaches are conventional, whereas the motif-based approach is novel in utilizing motif as a basic manipulation unit. Compared with previously published review papers on fabric inspection, this paper firstly offers an up-to-date survey of different defect detection methods and describes their characteristics, strengths and weaknesses. Secondly, it employs a wider classification of methods and divides them into seven approaches (statistical, spectral, model-based, learning, structural, hybrid, and motif-based) and performs a comparative study across these methods. Thirdly, it also presents a qualitative analysis accompanied by results, including detection success rate for every method it has reviewed. Lastly, insights, synergy and future research directions are discussed. This paper shall benefit researchers and practitioners alike in image processing and computer vision fields in understanding the characteristics of the different defect detection approaches.

Wavelet based methods on patterned fabric defect detection

The wavelet transform (WT) has been developed over 20 years and successfully applied in defect detection on plain (unpatterned) fabric. This paper is on the use of the wavelet transform to develop an automated visual inspection method for defect detection on patterned fabric. A method called direct thresholding (DT) based on WT detailed subimages has been developed. The golden image subtraction method (GIS) is also introduced. GIS is an efficient and fast method, which can segment out the defective regions on patterned fabric effectively. In this paper, the method of wavelet preprocessed golden image subtraction (WGIS) has been developed for defect detection on patterned fabric or repetitive patterned texture. This paper also presents a comparison of the three methods. It can be concluded that the WGIS method provides the best detection result. The overall detection success rate is 96.7% with 30 defect-free images and 30 defective patterned images for one common kind of patterned Jacquard fabric.

Novel method for patterned fabric inspection using Bollinger bands

This paper introduces a new application of Bollinger bands for defect detection of patterned fabric. A literature review on previous de- signed methods for patterned fabric defect detection will be depicted. For data analysis, Bollinger bands are calculated based on standard devia- tion and are originally used in the financial market as an oversold or overbought indicator for stock. The Bollinger bands method is an effi- cient, fast and shift-invariant approach, that can segment out the defec- tive regions on the patterned fabric with clear and crystal clean images. The new approach is immune of the alignment problem that often hap- pens in previous methods. In this paper, the upper band and lower band of Bollinger bands, which are sensitive to any subtle change in the input data, have been developed for use to indicate the defective areas in patterned fabric. The number of standard deviation and length of time of Bollinger bands can be easily determined to obtain excellent detection results. The proposed method has been evaluated on three different patterned fabrics. In total, 165 defect-free and 171 defective images have been used in the evaluation, where 98.59% accuracy on inspection has been achieved.

Performance Evaluation for Motif-Based Patterned Texture Defect Detection

This paper carries an extensive evaluation on the performance of a generalized motif-based method for detecting defects in 16 out of 17 wallpaper groups in 2-D patterned texture. The motif-based method evolves from the concept that every wallpaper group is defined by a lattice, which contains a further constituent-motif. It utilizes the symmetry properties of motifs to calculate the energy of moving subtraction and its variance among motifs. Decision boundaries are determined by learning the distribution of those values among the defect-free and defective patterns in the energy-variance space. In this paper, shape transform for irregular motif has been demonstrated according to the three basic motif shapes: rectangle, triangle, and parallelogram. An error analysis for the misclassifications has also been delivered. In the database of fabrics and other patterned textures, a total of 381 defect-free lattices are used for formulation of boundaries while further 340 defect-free and 233 defective lattices are for testing. The motif-based method has a consistent result and reaches a detection success rate of 93.86%.

Motif-based defect detection for patterned fabric

This paper proposes a generalized motif-based method for detecting defects in 16 out of 17 wallpaper groups in 2D patterned texture. It assumes that most patterned texture can be decomposed into lattices and their constituents-motifs. It then utilizes the symmetry property of motifs to calculate the energy of moving subtraction and its variance among different motifs. By learning the distribution of these values over a number of defect-free patterns, boundary conditions for discerning defective and defect-free patterns can be determined. This paper presents the theoretical foundation of the method, and defines the relations between motifs and lattice, from which a new concept called energy of moving subtraction is derived using norm metric measurement between a collection of circular shift matrices of motif and itself. It has been shown in this paper that the energy of moving subtraction amplifies the defect information of the defective motif. Together with its variance, an energy-variance space is further defined where decision boundaries are drawn for classifying defective and defect-free motifs. As the 16 wallpaper groups of patterned fabric can be transformed into three major groups, the proposed method is evaluated over these three major groups, from which 160 defect-free lattices samples are used for defining the decision boundaries, with 140 defect-free and 113 defective samples used for testing. An overall detection success rate of 93.32% is achieved for the proposed method. No other generalized approach can achieve this success rate has been reported before, and hence this result outperforms all other previously published approaches.

Ellipsoidal decision regions for motif-based patterned fabric defect detection

This paper presents a study of using ellipsoidal decision regions for motif-based patterned fabric defect detection, the result of which is found to improve the original detection success using max-min decision region of the energy-variance values. In our previous research, max-min decision region was found to be effective in distinct cases but ill detect the ambiguous false-positive and false-negative cases. To alleviate this problem, we first assume that the energy-variance values can be described by a Gaussian mixture model. Second, we apply k-means clustering to roughly identify the various clusters that make up the entire data population. Third, convex hull of each cluster is employed as a basis for fitting an ellipsoidal decision region over it. Defect detection is then based on these ellipsoidal regions. To validate the method, three wallpaper groups are evaluated using the new ellipsoidal regions, and compared with those results obtained using the max-min decision region. For the p2 group, success rate improves from 93.43% to 100%. For the pmm group, success rate improves from 95.9% to 96.72%, while the p4m group records the same success rate at 90.77%. This demonstrates the superiority of using ellipsoidal decision regions in motif-based defect detection.

Defect detection on patterned jacquard fabric

The techniques for defect detection on plain (unpatterned) fabrics have been well developed nowadays. This paper is on developing visual inspection methods for defect detection on patterned fabrics. A review on some defect detection methods on patterned fabrics is given. Then, a new method for patterned fabric inspection called Golden Image Subtraction (GIS) is introduced. GIS is an efficient and fast method, which can segment out the defective regions on patterned fabric effectively. An improved version of the GIS method using wavelet transform is also given. This research results contribute to the development of an automated fabric inspection machine for the textile industry.

Distance-based k-nearest neighbors outlier detection method in large-scale traffic data

This paper presents a k-nearest neighbors (kNN) method to detect outliers in large-scale traffic data collected daily in every modern city. Outliers include hardware and data errors as well as abnormal traffic behaviors. The proposed kNN method detects outliers by exploiting the relationship among neighborhoods in data points. The farther a data point is beyond its neighbors, the more possible the data is an outlier. Traffic data here was recorded in a video format, and converted to spatial-temporal (ST) traffic signals by statistics. The ST signals are then transformed to a two-dimensional (2D) (x, y) -coordinate plane by Principal Component Analysis (PCA) for dimension reduction. The distance-based kNN method is evaluated by unsupervised and semi-supervised approaches. The semi-supervised approach reaches 96.19% accuracy.

Patterned Fabric Defect Detection using a Motif-Based Approach

This paper proposed a patterned fabric defect detection method for sixteen out of seventeen wallpaper groups using a motif-based approach. From the symmetry properties of motifs, the energy of moving subtraction and its variance among motifs are mapped onto an energy-variance space. By learning the distribution of defect-free and defective patterns in this space, boundaries conditions can be determined for defect detection purpose. The proposed method is evaluated on four wallpaper categories, from which all 16 wallpaper groups can be generalized. Altogether, 160 defect-free lattices samples are used for learning the decision boundaries; and 200 other defect-free and 138 other defective samples are used for testing. An overall detection accuracy has reached 93.61%, which outperforms previous approaches.

Anomaly Detection for Quaternion-Valued Traffic Signals

In this paper, a novel anomaly detection method ispresented by using quaternion numbers to model traffic signals. A signal processing approach is proposed to deal with traffic surveillance. Traffic structures are depicted using directed graph models. The relationship among different traffic direction signals are represented through using quaternion numbers instead of individual representation of one particular direction. Multi- granularity local density-based method is adopted to perform anomaly detection for separate entry direction distribution (EDD) signals. Complex traffic signals are subsequently examined by exploring the relationship expressed with quaternion numbers. In such way, the anomaly detection complexity is reduced. Experimental results show that the proposed algorithm can achieve high detection rate. The overall average DSR of both AM and PM sessions is about 97.83%, which is better than the previous algorithm (96.67%) in the literature.