ABSTRACT : This study has shown that image analysis has great potential to provide reliable measurements for defect detection of knitted fabrics. Using the principles of image analysis, an automatic fabric evaluation system, which enables automatic computerized defect detection analysis of knitted fabrics, was developed. Online fabric defect detection was tested automatically by analyzing fabric images captured by digital camera. The results of the automatic fabric defect detection correspond well with the experimental values. Therefore, it is shown that the developed image capturing and analysis system is capable of on line fabric defect detection and full control in knitting machine (i.e., stopping the circular knitting machine as soon as defect acquisition at once by the digital camera.

INTRODUCTION :
Any variation to the knitting process needs to be investigated and corrected. Defects fall into this category. Since when they appear, repair is needed, which is time consuming and sometimes results in fabric rejection.Fabric defect detection has been a long felt need in the textile and apparel industry. Surveys carried out as early as 1975 [1] show that inadequate or inaccurate inspection of fabrics has led to fabric defects being missed out, which has in turn had great effects on the quality and subsequent costs of the fabric finishing and garment manufacturing processes.

Circular knitting is one of the easiest and fastest ways (20 million stitches per minute) of producing cloth and textile pieces such as garments, socks and gloves. Fabric faults, or defects, are responsible for nearly 85% of the defects found by the garment industry. An automated defect detection and identification system enhances the product quality and results in improved productivity to meet both customer demands and to reduce the costs associated with off-quality. Higher production speeds make the timely detection of fabric defects more important than ever. Presently, inspection is done manually when a significant amount of fabric is produced, the fabric roll is removed from circular knitting machine and then sent to an inspection frame. An optimal solution would be to automatically inspect fabric as it is being produced and to encourage maintenance personnel to prevent production of defects or to change process parameters automatically and consequently improve product quality [2].

The study of this problem has led to the identification of two main categories of defects in knitted fabrics: horizontal and vertical variations [3,4]. While the first category is mainly due to the yarn (quality and management), the second category is related with the knitting elements: needles, sinkers, feeders, and so on. The solutions encountered for solving these problems are, for the first category, a careful selection and management of the yarn, and for the second, the correction or substitution of the defective elements. In order to deal with these problems, various studies have been conducted and some specialised systems were developed, which can detect abnormalities in the yarn being fed, defects in the knitted fabric and defects in the knitting elements [4,5].

In the previous work [6], the neural network methods applied on images of simple circular knitted fabrics for classification of faults. The result showed the successfulness of the methods but this approach can not be useful for industries because the process is time consuming and there are no ways for determination of fault location and area. To prevent the problem, in current research [7], the wavelet transform was applied for processing of image of circular knitted fabrics.

Depending on knitted structure, the defects can be categorized in three types of vertical, horizontal and regional defects [8, 9].

Table (1): Nominal value for the extracted features of both standard (s) and defective fabrics (d) for single jersey using radon transform.

Table (2) shows nominal values for the statistical features, for (24) normal cases and
(24) defective ones for rib structure, using radon transform.

3.RESULTS:

The performance of the described inspection system was evaluated in seven stages using the simulated circular knitting machine. In the first stage, moving the camera-carrier in a reciprocating motion along the height of the simulated cylinder of the circular knitting machine with a slow speed (10 mm/ minute). In the same time, the simulated cylinder was rotated at the speed of 15 r.p.m as an operator introduced defects into the knitting process. The second stage of testing is the isolation of the mounting components from the considerable vibration that is produced during circular knitting machine operation. The third stage is defect acquisition and its analysis quickly at a rate of 10 images pre second with a high- quality. The fourth stage is the identification of defect type. The fifth stage is stopping the simulated circular knitting machine at once as soon as the defect is acquisited by the camera. The sixth stage is appearing the defect name, its picture, causes and remending method on the monitor of the computer to help the operator.And the seventh stage is the designing of an inspection system whose cost- effectiveness can justify its use on many, if not all, of the circular knitting machines in a knitting mill.The four defect types for both plain single jersey and rib structures included almost all of the most commonly occurring knitting defects, such as needle line, dropped stitch, hole and oil spots.

The image acquisition subsystem consistently produced high- quality images of the knitted fabric for both plain single jersey and rib structures. The image resolution was set at 320 pixels / 6.5 cm (125pixets/inch). The higher resolution was necessary because the impurities that are naturally present in the knitted-yarn fabric tend to obscure the more subtle defects.

Note that with a nominal circular knitting machine speed of 15 r.p.m and a maximum (125 pixels /inch) 320 pixels/6.5 cm, the exposure time of the digital camera is less than the shortest time between forward motion pulses (i.e., 0.25 sec) and therefore is sufficient to stop the motion of the machine. During the analysis of these images, the acquisition subsystem was directed to capture the next 320x240 image frame so that 100 % coverage of the knitted fabric was maintained.In analyzing more than 2000 for both fabric type, the overall defection rate of the presented approach was found to be 92% with a localization accuracy of 3 mm and a false alarm rate of 2.5%. The false alarm rat was computed as the total number of false detections divided by the total number of processed images. Note that the detection rate of 92% represents an average over all defect types. In general, because we are dealing with an edge-based segmentation approach defects that produce very subtle intensity transitions (e.g., mixed yarns and barre) were detected at a lower rate (i.e., 50-60%). On the other hand for the most commonly occurring and the most serious defects, such as needle line, dropped stitch, holes and oil spots, the defection rate was 92%.. Because some times the camera captures a part of the defect but not all, therefore the defect can be classified with another type for example dropped stitch defect can be shown as a hole. Matlab has been chosen as the programming language in which to develop the software for the purpose of this research. It is intended to use a back propagation network for the processing. Table (1) shows nominal values for the statistical features, for (30) normal cases and (30) defective ones for single jersey, using radon transform.

Table (2): Nominal value for the extracted features of both standard (s) and defective fabrics (d) for rib structure using radon transform.
A number of simple heuristic shape descriptors (texture features)for single jersey is exist in Table (3).
Table (3) : Heuristic Shape Descriptors(texture features) for single jersey.

A number of simple heuristic shape descriptors (texture features)for rib structure is exist in Table (4).
Table (4) : Heuristic Shape Descriptors(Texture Features) for Rib Structure.

Figures (8,9) and Tables (5,6) show the training pattern of the used network using (60,48) different samples (normal & defected) for both single jersey and rib structures.

Fig (8): RMS error versus learning count of the used network for single jersey.

Fig (9): RMS error versus learning count of the used network for rib structure.

Table (5):Classification results for the training patterns for single jersey.

Table(6):Classification results for the training patterns for rib structure .

4.CONCLUSION:

We have described a computer vision aided fabric inspection System to detect and classify circular knitted fabric defects using common different texture recognition methods including, thresholding analysis. Radon transform, discrete fourier transform and neural network. It was found that the application of discrete fourier transform method applied in this work is highly promising in the identification of knitted fabric defects with the overall success rate of 92% has the highest efficiency value among the other methods.

The results gained from these experience show that discrete fourier transform act precisely and fast for defecting faults and specifying their area, as well as being used as on- line detection tool in knitting machines during knitted fabric reduction. Finally usage of such a process can eliminate further human inspection stage. In addition, the circular knitting machine could be controlled and stopped at once as soon as the defect is captured by the camera. We have described a vision- based fabric inspection system that accomplishes on circular knitting machine inspection of the knitted fabric with 100% coverage. The inspection system is scalable and can be manufactured at relatively low cost using off- the shelf components. This system differs from those reported in the literature in two crucial ways. First, it is on- circular knitting machine; and second, it is equipped with a novel defect segmentation technique, which was thoroughly tested under realistic conditions and was found to have a high detection rate and accuracy and a low rate of false alarms. The fabric inspection system texture was described in terms of its image acquisition subsystem and its defect segmentation algorithm. The image acquisition subsystem is used to capture high-resolution, vibration-free images of the knitted fabric under construction. The essence of the presented segmentation algorithm is the localization of those defects in the input images that disrupt the global homogeneity of the background texture. Novel texture features are utilized to measure the global homogeneity of the output images. A prototype system was used to acquire and to analyze more than 2000 images of fabrics that were constructed with two different types of knitted structure. In each case, the performance of the system was evaluated as an operator introduced defects from 14 categories into the knitting process. The overall defection rate of the presented approach was found to be 92% with a localization accuracy of 3 mm and a false alarm rate of 2.5%.

REFERENCES:

1- Knell, A. L., Automatic Fabric Inspection, Textile Institute and Industry, January, 1975.
2-Dorrity J. L., Vachtsevanos G. and Jasper W., Real-time Fabric Defect Detection and Control in Weaving Processes, National Textil Center Annual Report, November, (1996).
3- "Les Defaults des Tricots", Centre D'Etude et de Recherche de la Maille, ITF Maille, Troys.
4- De Araujo, M. D., "Manual de Engenharia Textil Vol. 1 (1986) Lisboa: Fundacao Calouste Gulbenkian.
5-"Reglage Rationnel des Metiers Circulaires", ENSITM, Universite de Haute Alsace, Mulhouse.
6- Salamaty G., "Classification of Circular Woven Fabrics Using Neural Networks", M. Sc. Project, Amirkabir University of Technology, 2001.
7- Sadrei A. H., "Fault Detection on Simple Circular Knitted Fabrics Using Wavelet", M. Sc. Project, Amirkabir University Technology, 2001.
8- Goldberg S., "Knitted Fabric Technology", National Knitted Outerwear Association, 1974.
9- Iyer C., Mammel B. and Schach W., "Circular Knitting", Meisenbach Bamberg, 1995.
Elsevier Science Ltd. Great Britain, 1995.
10- Chrpova E., Hotar V. and Lang M., "Application of "Noviscam Technique", 1st ISQVPFD, Bled, Slovenia, 2000.
11- Chrpova E., Surface Quality Control Based on Image Processing Methods. Fibres and Textiles Vol. 10 (2), Bratislava, Slovak Republic, 2003.
12- Chrpova E., Trmal G. J. and Hotar V., A fast Image Processing Method for Quality Control of Textiles. Tehnicki Vjesnik / Technical Gazette Vol. 10 (3/4), Slavonski Brod, Croatia, 2003.
13- Ribolzi, S., Merckle, J. Gresser, J. and Exbrayat, P. E., "Real-Time fault detection on textiles using Opto-electronic processing" Textile Res. J. 63(2), 61-71 (1993).
14- Shou Tsai, I., Chung-Hua Lin, and Jeng-Jong Lin, "Applying an Artificial Neural Network to pattern recognition in fabric defects". Textile Res. J. 65(3), 123-130 (1995).

15- Warren J. Jasper and Harsh Potlapalli, "Image analysis of mispicks in woven fabric" Textile Res. J., 65(11), 683-692, (1995).
16-Bugao Xu, "Identifying fabric structures with Fast Fourier Transform techniques", Textile Res. J. 66(8), 496-506 (1996).
17-Alam Eldin A. T. and Nour Eldin H. A., "Automated visual inspection of web-type products using unitary transforms" the 36th ISMM International Conference, Mini and Microcomputers and their applications, MIMI 88, June 27-30, Barcelona, SPAIN, 1988.
18-Hassan H. Soliman, "Theory and applications of Hardware-Implemented algorithms" Ph. D. Thesis, Mansoura University, EGYPT, 1992.
19-Shou Tsai, I., and Ming-Chuan Hu, "Automatic inspection of fabric defects using an Artificial Neura1 Network technique " Textile Res. J. 66(7), 474-482 (1996).
20-Ebraheem Shady, "A computer vision system for automated inspection of fabrics" Master Thesis. Mansoura University, Egypt (1998).
21-http://fibers.philau.edu/ntc/SO1PH07/.
22-Araujo, M. D., Catarino, A. and Hong, H., Process Control for Total Quality in Circular Knitting, AUTEX Research Journal, Vol 1, No.l, (1999).
23-Araujo, M. D., Manual de Engenharia Textil, Vol 1, Lisboa: Fundacao Calouste Gulbenkian, (1986).
24-Les Defaults des Tricots, Centre D'Etude et de Recherche de la Maille, ITF Maille, Troys.
25-Rsglage Rationnel des Metiers Circulaires, ENSITM, Universite de Haute Alsace, Mulhouse.
26-Tsai, D. M. and Heish, C. Y., Automated surface inspection for directional textures, Image and Vision Computing, vol. 18, pp. 49-62, (1999).


To read more articles on Textile, Fashion, Apparel, Technology, Retail and General please visit www.fibre2fashion.com/industry-article

To promote your company, product and services via promotional article, follow this link: https://www.fibre2fashion.com/services/featrued-article/featured_article.asp

This paper presents a method and instruments developed at Textile Faculty for the automatic detection and identification of defects on flat fabrics. The work was intended to develop monitoring system for random processes based on video image during the production phase. Our developed system consists of hardware equipment [10-12], data evaluation implemented in software and determination of acceptable tolerances related to final product quality. There were investigated applications of the developed methods in a paper production process, a carding process and a weaving process.

Many researchers in the field of image analysis used neural network as a classifier [13-18]. In these approaches, the data of the images is reduced, in one form or another, to accelerate the processing time. Techniques used to extract image features include statistical procedures [14, 16, 19], time-frequency domain transforms such as Discrete Cosine Transform [14,19], Fourier Transform [14-16,20,21,22] and Wavelet Transform [21].

It has been reported [18] that knitted fabric defects can be classified in two main categories including horizontal and vertical variations [23,24]. While the first category is mainly due to the yarn, the second category is related to the knitting elements. In order to deal with these problems, various studies have been conducted and some specialized systems were developed, which can detect abnormalities in the yarn being fed, defects in the knitted fabric and defects in the knitting elements [23,25]. It was claimed that these systems are very specialized and usually do not give further information related with the knitting process and the cause of a defect.

The objective of this research is the development of a computerized system capable of detecting defects in knitted fabrics during the knitting process. Further, this system should be able to identify the type and potential source of the defect, providing the operator with information about how to correct the problem. The developed systems are capable of identifying defects with greater accuracy than experts in the knitting industry, promising significant improvements in quality. In addition, this system is capable to stop the circular knitting machine as soon as defect acquisition at once.

2. EXPERIMENTAL WORK:

2-1 Sample and Tested Defects:


In this study, we used two types of fabric structure such as plain single jersey and rib with different densities. With respect to the tested fabrics, four kinds of common defects were chosen and created in these fabrics. These defects are needle line, dropped stitch, hole and oil spots. From each of the three kinds of fabric defects, a lot of samples were acquired by image-capture equipment.

Each type of defects is imaged many times from random different locations. The images of these defected samples were analyzed by using a computer program.

The samples for each kind of fabric defect were divided into two groups. The first group had many samples and was used for training , and the other group was used for testing.

2-2- Fabric Image Acquisition:

The experimental materials include plain single jersey and rib structures, twenty specimens for each weave pattern acquired from different areas of the same knitted fabrics. The illumination consists of a fluorescent lamp, it was inclined 45o to the specimen surface 5 to 13cm apart and the magnification was 6 x. The captured images of knitted fabrics consist of RGB 24 (320 x 240) pixels and each pixel has 256 gray levels. The area of a knitted fabric sample was 6,5x4,8cm.

On- circular knitting machine fabric image acquisition presents several challenges to obtaining high quality, high resolution images. One of these challenges is the isolation of the mounting components from the considerable vibration that is produced during circular knitting machine operation. Another challenge is the designing of an inspection system whose cost-effectiveness can justify its use on many, if not all, of the circular knitting machines in the knitting mill.

As described in the following sections, each of these challenges has been addressed and met in developing our (on-circular knitting machine) image acquisition subsystem.

A - Hardware Description:

The image acquisition subsystem is implemented with standard components on a low-cost personal computer. These components, shown in Fig.(1), consist of a 80 - elements, simulated circular knitting machine, digital camera, a source of illumination for front lighting the fabric , interface board, parallel port , personal computer (PC) and display monitor.

These components are used to acquire high resolution, vibration-free images of the fabric under construction and to store them on the personal computer memory. The software running on the interface board controls the image acquisition process and accumulates a two dimensional (2-D) image suitable for the ensuing analysis (i.e., defect segmentation).

B- Image Acquisition Operation:

During image acquisition, the camera exposure time is designed to be fixed, regardless of the sumulated circular knitting machine speed. The fixed exposure time is realized by the exposure time control of the camera-encoder interface.

The acquired image frame serves as an input to the image analysis or, more specifically, to the defect segmentation algorithm, which is also executed on the interface board. To maintain full coverage of the fabric, the acquisition subsystem begins capturing the next frame while the current frame is analyzed for defects; the following section presents a detailed description of the defect segmentation algorithm.

2-3- Defect Segmentation Algorithm:

In designing the defect segmentation algorithm for our inspection system, we observed that the images of fabrics constitute ordered textures that are globally homogenous, that is, statistics measured from different patches in an image are correlated. It was further noted that images containing defects are less homogenous than those that are defect- free. Therefore, the essence of the presented segmentation algorithm is to localize those events (i.e., defects) in the image that disrupt the global homogeneity of the background texture. We shall now describe the algorithmic modules as shown in Fig. (2) that are designed to accomplish this very goal under the following conditions:

1- Defects exhibit low-intensity variation within their boundary
and
2- Relative to the textured background, they constitute a small portion of field of the field of view.

Acquired images are transferred to the host computer and processed by the procedures in Fig.(2).

In the following sections, the modules are described in detail and their efficacy is clearly demonstrated using the images captured by the image acquisition subsystem.

D-Feature Extraction System:

Having preprocessed the digitized mammograms and isolated the knitting defects from its background, the next step is to extract some features having the ability to discriminate between normal (standard) and defective fabrics. These features will be used as the input to the classifier.

This chapter introduces a detailed discussion of the feature extraction stage. Two sets of features are introduced; statistical features and texture features.

i - Statistical Features:

The main interest of feature extraction is to find some characteristics that distinguish between normal (standard) and defective fabrics. Fig (5) shows several pairs of histograms, for each pair the first histogram is for a normal knitted fabric and the second is for defective one. It is obvious from the figures that there is a discriminating difference between normal histograms and defective ones. The histograms of the normal cases are unimodal and the grey level values are centered around the mean values with small variance, while those of defective cases are bimodal or multimodal. The grey level values are spreaded over a wider range of values.

Statistical features are numerical descriptive measure of the histogram, knowledge of parameters allows us to reduce large amounts of information into a summary form that is easy to interpret. These parameters describe the states of nature in decision problem [26]. Therefore, statistical features like mean, standard deviation,variance, coefficient of variation, moment, skewness and kurtosis are used to characterize the histograms and to distinguish between normal and defective fabrics. The mathematical definitions of these features are:

where N is the number of pixels in the sub-image.
E is the expected value of X.

ii  Texture Features:

We can use boundary information to describe a region, and shape can be described from the region itself A large group of shape description techniques is represented by heuristic approaches which yield acceptable results in description of simple shapes. Region area, rectangularity, elongatedness, direction, compactness, etc., are examples of these methods. Unfortunately, they cannot, be used for region reconstruction and do not work for more complex shapes.