FABRIC DEFECT DETECTION WITH ARTIFICIAL

INTELLIGENCE WITH REGION FEATURES

ABSTRACT:

The development of machine vision technology, artificial vision is

gradually replaced by machine vision, which avoids most of the
errors caused by artificial detection, and correspondingly improves
the detection accuracy and efficiency. At present, domestic and
foreign researchers are committed to applying deep learning
technology to defect detection of various fabrics and cloths. The
fabric image was processed by using the lower convolution feature
layer to detect small target defects. This system focused on the
innovative detection of fabric defects, a precise fabric defect
detection system based on Artificial neural network and computer-
aided is designed and developed, which can achieve high detection
accuracy. A series of improvement measures are carried out in the
traditional machine learning algorithm. Based on the region features
of the target, we train the network with features from Expectation
maximization segmented Image. For the complex features of the
target, we use Adaptive neural network with radial basis fucntion.
INTRODUCTION:

In industrial production, product quality inspection is particularly

important. At present, the most commonly used method for surface
defect detection of industrial products is mainly manual inspection.
However, manual defect inspection is easily affected by the subjective
factors and work experience of the inspector, and the accuracy rate is
not high. When the defect is small, the human eye cannot detect it,
and it’s not suitable for manual inspection when some inspection
environments are harmful to the human body. In recent years, surface
defect detection of industrial products based on visual perception has
gradually replaced manual defect detection and become an
indispensable part of industrial product surface defect detection.
Surface defect detection of industrial products based on visual
perception technology is a non-contact automatic detection
technology. The widespread application of surface defect detection of
industrial products based on visual perception not only reduces labor
costs, but also significantly improves production efficiency. Surface
defect detection of industrial products based on visual perception can
be mainly divided into traditional image processing-based industrial
product surface defect detection and deep learning-based industrial
product surface defect detection. Traditional image processing-based
industrial product surface defect detection requires manually designed
features. This method of manual feature design avoids using a large
number of data to learn features, but it has stricter requirements on the
imaging environment and poor adaptability. Deep learning-based
industrial product surface defect detection needs to automatically
extract features from a large number of data. This method of
automatic feature extraction is highly adaptable to the environment,
but requires a large number of training data. In the real industrial
environment, due to the low probability of occurrence of defective
samples, it will be difficult to collect defective samples, and defect
labeling requires a lot of manpower and material resources. Therefore,
surface defect detection of industrial products based on a small
number of labeled data is particularly important. This paper
introduces the method of surface defect detection of industrial
products based on a small number of labeled data, and the method is
roughly divided into traditional image processing-based industrial
product surface defect detection method and deep learning-based
industrial product surface defect detection methods suitable for a
small number of labeled data. The structure of this paper is as follows.
Section 2 introduces traditional image processing-based industrial
product surface defect detection methods, which are divided into
statistical methods, spectral methods and model methods. Section 3
introduces machine learning-based industrial product surface defect
detection methods suitable for a small number of labeled data, which
are divided into based on data augmentation, based on machie
learning, model-based fine-tuning, semi-supervised, weak supervised
and unsupervised.
FABRIC DEFECTS

It has been estimated that the price of fabrics is reduced by 45%-65%

due to the presence of defects. The fabric quality is affected by yarn
quality and/or loom defects. The poor quality of raw materials and
improper conditioning of yarn result in yarn quality defects and
effects such as color or width inconsistencies, hairiness, slubs, broken
ends, etc.. There are numerous quality tests for yarns, such as ASTM
D2255-96 , for predicting the quality of fabric to be produced from
the entire lots of sampled yarns. The tests on the quality of yarns are
usually performed at the output of spinning-mills. Quality test runs for
looms and knitting machines require interruption of the weaving
process . This interruption is not practically feasible for the machines
that are intended for large production runs of fabric rolls. The quality
test runs on the older, worn, or obsolete model weaving machines
generally produce unacceptable results. These test runs tend to be
smaller and may not register recurring fabric defects that are
generated due to sinusoidaly occurring inconstancies in the weaving
machines. Therefore, such fabric defects can be incorrectly read as
resulting from poor yarn quality. The fabric defects resulting from
variations in the tension of one or more yarn strands are generally
misread as the defects resulting from poor yarn quality. The weaving
irregularities generated in the weaving machines due to the change in
operating conditions (temperature, humidity, etc.) also result in
various fabric defects independently of yarn quality. The population
of fabric defects may vary dynamically as small changes in the
weaving process can result in entirely new class of fabric defects

Traditional inspection

There are two distinct possibilities for fabric defect detection. The
first possibility is the process inspection in which the weaving process
(or its parameters) can be constantly monitored for the occurrence of
defects. Process inspection is a preventive inspection, and is generally
not performed in the textile industries due to the complexity of the
weaving process. The second possibility is the product (end)
inspection in which the manufactured fabric has to be inspected for
the defects. The present research is focused on product inspection.
The fabric produced from the weaving machines is about 1.5-2 meters
wide, and rolls out at the speed of 0.3-0.5 meters per minute. The
product inspection in the textile industries is not performed
concurrently with the production. The slow speed of manufactured
fabric is insufficient to keep a human inspector occupied and human
inspection is therefore uneconomical. Also, the relatively hostile
working environment near the weaving machines is not suitable for
human inspection. The traditional inspection procedure is to remove
the manufactured fabric rolls from the weaving machines and unroll
them on the inspection table (specially illuminated) at a relatively
higher speed of 8-20 meters per minute. When a human inspector
notices a defect on the moving fabric, he stops the motor that moves
the fabric roll, records the defect and its location, and starts the motor
again. The early detection of repetitive defects and extraordinary
defect rate is left to the operators or so called roving inspectors. These
roving inspectors will warn the production department so that
appropriate measures can be taken to decrease the defect rate.

Fabric defect detection is a quality control process that has to ensure

the identication of defects present in the textile fabric. These defects
can reduce the textile fabric price as much as 45% to 65% [1]. A
traditional inspection system is composed of manual
workers/operators. Their job is to detect the defects while the fabric is
being moved by a machine.Traditionally, this motorized machines
unroll the fabric rolls, so that the fabric is stretched and presented to
the worker without folds and thickness differences. As this process
relies on human visual ability and concentration, this task can be very
tedious and time-consuming, which can lead to fatigue and
consequently human error. Therefore, traditional systems can only
achieve a 60% to 75% accuracy, even though they have very slow
speed compared to production rate. As a result, automatic visual
inspection systems to ensure the high quality of products in
production lines are in increasing demand.
LITERATURE REVIEW:

Xiang Jun , Jingan Wang et al proposed With the rise of

labor costs and the advancement of automation in the textile
industry, fabric defect detection has become a hot research field
in recent years. We proposed a learning-based framework for
automatic detection of fabric defects. Firstly, we use a fixed-size
square slider to crop the original image to a certain step and
regularity. Then an improved histogram equalization is used to
enhance each cropped image. Furthermore, the Inception-V1
model is employed to predict the existence of defects in the
local area. Finally, we apply the LeNet-5 model, which plays the
role of a voting model, to recognize the type of the defect in the
fabric.
Maheshwari S. Biradara, *, B. G. Shiparamattia et al
proposed The enormous growth in the fashion industry
increased the demand for quality of service of the fabric
material. Fabric defect detection plays a crucial role in
maintaining the quality of service as a single defect in the fabric
can halve its price. Traditional machine learning approaches are
less generalized and cannot be employed for fabric defect
detection of patterned as well as non-patterned fabrics. This
paper presents Deep Convolutional Neural Network (DCNN)
for fabric defect detection. The proposed method consists of a
three-layered DCNN for the representation of the normal and
defected fabric patch.
A Survey of Surface Defect Detection of Industrial
Products Based on A Small Number of Labeled Data
Qifan Jin (1), Li Chen et al proposed The surface defect
detection method based on visual perception has been widely
used in industrial quality inspection. Because defect data are not
easy to obtain and the annotation of a large number of defect
data will waste a lot of manpower and material resources.
Therefore, this paper reviews the methods of surface defect
detection of industrial products based on a small number of
labeled data, and this method is divided into traditional image
processing-based industrial product surface defect detection
methods and deep learning-based industrial product surface
defect detection methods suitable for a small number of labeled
data

Yubo Huang and Zhong Xiang et al proposed On a global

scale, the process of automatic defect detection represents a
critical stage of quality control in textile industries. In this paper,
a semantic segmentation network using a repeated pattern
analysis algorithm is proposed for pixel-level detection of fabric
defects, which is termed RPDNet (repeated pattern defect
network). Specifically, we utilize a repeated pattern detector
based on convolutional neural network (CNN) to detect periodic
patterns in fabric images. Through the acquired repeated pattern
information and proper guidance of the network in a high-level
semantic space, the ability to understand periodic feature
knowledge and emphasize potential defect areas is realized.
Concurrently, we propose a semi-supervised learning scheme to
inject the periodic knowledge into the model separately, which
enables the model to function independently from further pre-
calculation during detection, so there is no additional network
capacity required and no loss in detection speed caused. In
addition, the model integrates two advanced architectures of
DeeplabV3+ and GhostNet to effectively implement lightweight
fabric defect detection
Ajaykumar et al proposed The investment in automated
fabric defect detection is more than economical when reduction
in labor cost and associated benefits are considered. The
development of fully automated web inspection system requires
robust and efficient fabric defect detection algorithms. The
inspection of real fabric defects is particularly challenging due
to the large number of fabric defect classes which are
characterized by their vagueness and ambiguity. Numerous
techniques have been developed to detect fabric defects and the
purpose of this paper is to categorize and/or describe these
algorithms. This paper attempts to present the first survey on
fabric defect detection techniques presented in about 160
references. Categorization of fabric defect detection techniques
is useful in evaluating the qualities of identified features. The
characterization of real fabric surfaces using their structure and
primitive set has not yet been successful. Therefore on the basis
of nature of features from the fabric surfaces, the proposed
approaches have been characterized into three categories;
statistical, spectral and model-based. In order to evaluate the
state-of-the-art, the limitations of several promising techniques
are identified and performances are analyzed in the context of
their demonstrated results and intended application.
EXISTING WORK:

Quality control is an area of utmost importance for fabric production

companies. By not detecting the defects present in the fabrics,
companies are at risk of losing money and reputation with a damaged
product. In a traditional system, an inspection accuracy of 60-75% is
observed. In order to reduce these costs, a fast and automatic defect
detection system, which can be complemented with the operator
decision, is proposed in this paper. To perform the task of defect
detection, a custom Convolutional Neural Network (CNN) was used
in this work. Additionally, as an undetected defect (False Negative -
FN) usually has a higher cost to the company than a non-defective
fabric being classified as a defective one (false positive), FN
reduction methods were used in the proposed system

DRAWBACKS

They does not concentrate on defected area on the materials. They

achieves low prediction performance of Fabric Defect prediction.
PROPOSED SYSTEM:

This system focused on the innovative detection of fabric defects, a

precise fabric defect detection system based on convolution neural
network and computer-aided is designed and developed, which can
achieve high detection accuracy. A series of improvement measures
are carried out in the traditional machine learning algorithm. Based on
the region features of the target, we train the network with features
from Expectation maximization segmented Image. For the complex
features of the target, we use Adaptive neural network with radial
basis fucntion.
MODULES:
1). Image Acquisition
2). Pre processing
a). Resizing
b). Color Conversion
c). Noise Removal
3). Segmentation using Expectation Maximisation algorithm
4). Feature Extraction
a). Statistical based feature
b). Texture based feature
c). Shape based feature
5). Classification using ANN
6). Defect and Nondefect fabric
FLOW CHART

Image
Acquisition

Pre processing

Resizing

Color
Conversion

Noise Removal

Expectation
Maximisation
Segmentation

Feature Extraction

a. Statistical
b. Texture

Adaptive Neural Network with

radial Basis function

Normal Abnormal
Module Description:

Image Acquisition:
Image Acquisition is a process of getting an input image for the
process of automatic detection of material Defects using Digital
Image Processing.

Pre processing:
Pre-processing is a common name for operations with images at
the lowest level of abstraction both input and output are
intensity images. The aim of pre-.processing is an improvement of
the image data that suppresses unwanted distortions or enhances
some image features important for further processing.

Noise Removal using Median Filter:

The median filter is a nonlinear digital filtering technique, often used
to remove noise from an image or signal. Such noise reduction is a
typical pre-processing step to improve the results of later processing
(for example, edge detection on an image). Median filtering is very
widely used in digital image processing because, under certain
conditions, it preserves edges while removing noise (but see
discussion below), also having applications in signal processing. The
main idea of the median filter is to run through the signal entry by
entry, replacing each entry with the median of neighboring entries.

Segmentation:

Digital image processing is the use of computer algorithms to

perform image processing into digital images. Image segmentation is
very important and challenging process of image processing. Image
segmentation is the techniques are used to partition an image into
meaningful parts have similar features and properties. The aim of
segmentation is simplification i.e. representing an image into
meaningful and easily analyzable way. Image segmentation is the first
step in image analysis. The main goal of image segmentation is to
divide an image into several parts/segments having similar features or
attributes. The main applications of image segmentation are: Medical
imaging, Content-based image retrieval, and Automatic traffic control
systems, Object detection and Recognition Tasks, etc. The image
segmentation can be classified into two basic types: Local
segmentation (concerned with specific part or region of image) and
Global segmentation (concerned with segmenting in whole image,
consisting of large number of pixels). We use Expectation
Maximisation Segmentation for extraction of Pneumonia.

Expectation Maximisation:
The EM algorithm proceeds from the observation that there is a way
to solve these two sets of equations numerically. One can simply pick
arbitrary values for one of the two sets of unknowns, use them to
estimate the second set, then use these new values to find a better
estimate of the first set, and then keep alternating between the two
until the resulting values both converge to fixed points. It's not
obvious that this will work, but it can be proven that in this context it
does, and that the derivative of the likelihood is (arbitrarily close to)
zero at that point, which in turn means that the point is either a
maximum or a saddle point.[13] In general, multiple maxima may
occur, with no guarantee that the global maximum will be found.
Some likelihoods also have singularities in them, i.e., nonsensical
maxima. For example, one of the solutions that may be found by EM
in a mixture model involves setting one of the components to have
zero variance and the mean parameter for the same component to be
equal to one of the data points.

The EM algorithm (and its faster variant ordered subset expectation

maximization) is also widely used in medical image reconstruction,
especially in positron emission tomography and single photon
emission computed tomography. See below for other faster variants of
EM.

In structural engineering, the Structural Identification using

Expectation Maximization (STRIDE) algorithm is an output-only
method for identifying natural vibration properties of a structural
system using sensor data
Feature Extraction:

Feature extraction from images involves extracting a minimal

set of features containing a high amount of object or scene
information from low-level image pixel values, therefore, capturing
the difference among the object categories involved. Some of the
traditional feature extraction techniques used on images are Scale-
invariant feature transform (SIFT) , histogram of oriented gradients
(HOG) , Local binary patterns (LBP), Content-Based Image Retrieval
(CBIR) ,etc. Once features are extracted their classification is done
based on objects present in an image. A few examples of classifiers
are Support vector machine (SVM), Logistic Regression, Random
Forest, decision trees etc

Statistical Features:
Mean:

For a data set, the arithmetic mean, also called the mathematical
expectation or average, is the central value of a discrete set of
numbers: specifically, the sum of the values divided by the number of
values. The arithmetic mean of a set of numbers x1, x2, ..., xn is

typically denoted by , pronounced "x bar". If the data set were

based on a series of observations obtained by sampling from
a statistical population, the arithmetic mean is the sample mean to
distinguish it from the mean of the underlying distribution,
the population mean . In probability and statistics, the
population mean, or expected value, are a measure of the central
tendency either of a probability distribution or of the random
variablecharacterized by that distribution.[2] In the case of a discrete
probability distribution of a random variable X, the mean is equal to
the sum over every possible value weighted by the probability of that
value; that is, it is computed by taking the product of each possible
value x of X and its probability p(x), and then adding all these

products together, giving .[3] An analogous formula applies to

the case of a continuous probability distribution. Not every
probability distribution has a defined mean; see the Cauchy
distribution for an example. Moreover, for some distributions the
mean is infinite.

For a finite population, the population mean of a property is equal to

the arithmetic mean of the given property while considering every
member of the population. For example, the population mean height
is equal to the sum of the heights of every individual divided by the
total number of individuals. The sample mean may differ from the
population mean, especially for small samples. The law of large
numbers dictates that the larger the size of the sample, the more likely
it is that the sample mean will be close to the population mean.[4]

Standard Deviation:

In statistics, the standard deviation (SD, also represented by the

lower case Greek letter sigma σ or the Latin letter s) is a measure that
is used to quantify the amount of variation or dispersion of a set of
data values.[1] A low standard deviation indicates that the data points
tend to be close to the mean (also called the expected value) of the set,
while a high standard deviation indicates that the data points are
spread out over a wider range of values.

The standard deviation of a random variable, statistical

population, data set, or probability distribution is the square root of
its variance. It isalgebraically simpler, though in practice less robust,
than the average absolute deviation. A useful property of the standard
deviation is that, unlike the variance, it is expressed in the same units
as the data.

In addition to expressing the variability of a population, the standard

deviation is commonly used to measure confidence in statistical
conclusions. For example, the margin of error in polling data is
determined by calculating the expected standard deviation in the
results if the same poll were to be conducted multiple times. This
derivation of a standard deviation is often called the "standard error"
of the estimate or "standard error of the mean" when referring to a
mean. It is computed as the standard deviation of all the means that
would be computed from that population if an infinite number of
samples were drawn and a mean for each sample were computed.

It is very important to note that the standard deviation of a population

and the standard error of a statistic derived from that population (such
as the mean) are quite different but related (related by the inverse of
the square root of the number of observations). The reported margin
of error of a poll is computed from the standard error of the mean (or
alternatively from the product of the standard deviation of the
population and the inverse of the square root of the sample size,
which is the same thing) and is typically about twice the standard
deviation—the half-width of a 95 percent confidence interval.

Variance:
In probability theory and statistics, variance is the expectation of the
squared deviation of a random variablefrom its mean. Informally, it
measures how far a set of (random) numbers are spread out from their
average value. Variance has a central role in statistics, where some
ideas that use it include descriptive statistics,statistical
inference, hypothesis testing, goodness of fit, and Monte Carlo
sampling. Variance is an important tool in the sciences, where
statistical analysis of data is common. The variance is the square of
the standard deviation, the second central moment of a distribution,
and the covariance of the random variable with itself.

GLCM Feature:

The Gray Level Co-occurrence Matrix1 (GLCM) and associated

texture feature calculations are image analysis techniques. Given an
image composed of pixels each with an intensity (a specific gray
level), the GLCM is a tabulation of how often different
combinations of gray levels co-occur in an image or image section.
Texture feature calculations use the contents of the GLCM to give a
measure of the variation in intensity (a.k.a. image texture) at the
pixel of interest.

Echoview offers a GLCM Texture Feature operator that produces

a virtual variable which represents a specified texture calculation on
a single beam echogram.

Algorithm
The virtual variable is created in the following way (using the
settings on the GLCM Textre page of the Variable properties dialog
box identified in bold):

1. Quantize the image data. Each sample on the echogram is

treated as a single image pixel and the value of the sample is the
intensity of that pixel. These intensities are then further
quantized into a specified number of discrete gray levels as
specified under Quantization.

2. Create the GLCM. It will be a square matrix N x N in size

where N is the Number of levels specified under Quantization.
The matrix is created as follows:

a. Let s be the sample under consideration for the calculation.

b. Let W be the set of samples surrounding sample s which

fall within a window centered upon sample s of the size
specified under Window Size.

c. Considering only the samples in the set W, define each

element i,j of the GLCM as the number of times two
 samples of intensities i and j occur in specified Spatial
relationship (where i and j are intensities between 0
and Number of levels-1) .

The sum of all the elements i, j of the GLCM will be the

total number of times the specified spatial relationship
occurs in W.

d. Make the GLCM symmetric:

i. Make a transposed copy of the GLCM

ii. Add this copy to the GLCM itself

This produces a symmetric matrix in which the

relationship i to j is indistinguishable for the
relationship j to i (for any two intensities i and j). As
a consequence the sum of all the elements i, j of the
GLCM will now be twice the total number of times
the specified spatial relationship occurs in W (once
where the sample with intensity i is the reference
sample and once where the sample with intensity j is
the reference sample), and for any given i, the sum of
all the elements i, j with the given i will be the total
number of times a sample of intensity i appears in the
specified spatial relationship with another sample.

e. Normalize the GLCM:

 i. Divide each element by the sum of all elements

The elements of the GLCM may now be considered

probabilities of finding the relationship i, j (or j, i)
in W.

Calculate the selected Feature. This calculation uses only the

values in the GLCM. See:

 Energy

 Entropy

 Contrast

 Homogeneity

 Correlation

 Shade

 Prominence

The sample s in the resulting virtual variable is replaced by the

value of this calculated feature.

Software Used:

MATLAB:

MATLAB (matrix laboratory) is a multi-paradigm numerical

computing environment and proprietary programming
languagedeveloped by MathWorks. MATLAB
allows matrix manipulations, plotting of functions and data,
implementation ofalgorithms, creation of user interfaces, and
interfacing with programs written in other languages, including C, C+
+, C#, Java,Fortran and Python.

Although MATLAB is intended primarily for numerical computing,

an optional toolbox uses the MuPAD symbolic engine, allowing
access to symbolic computing abilities. An additional
package, Simulink, adds graphical multi-domain simulation
and model-based design for dynamic and embedded systems.

As of 2018, MATLAB has more than 3 million users worldwide.

MATLAB users come from various backgrounds
ofengineering, science, and economics.

Syntax:

The MATLAB application is built around the MATLAB scripting

language. Common usage of the MATLAB application involves using
the Command Window as an interactive mathematical shell or
executing text files containing MATLAB code.

Sturctures:

MATLAB has structure data types.[15] Since all variables in MATLAB

are arrays, a more adequate name is "structure array", where each
element of the array has the same field names. In addition, MATLAB
supports dynamic field names[16] (field look-ups by name, field
manipulations, etc.). Unfortunately, MATLAB JIT does not support
MATLAB structures, therefore just a simple bundling of various
variables into a structure will come at a cost.[
Functions:

When creating a MATLAB function, the name of the file should

match the name of the first function in the file. Valid function names
begin with an alphabetic character, and can contain letters, numbers,
or underscores. Functions are often case sensitive.

Interfacing with other language:

MATLAB can call functions and subroutines written in the

programming languages C or Fortran.[26] A wrapper function is
created allowing MATLAB data types to be passed and
returned. MEX files (MATLAB executables) are the dynamically
loadable object files created by compiling such functions. [27][28] Since
2014 increasing two-way interfacing withPython was being added.

Libraries written in Perl, Java, ActiveX or .NET can be directly called

from MATLAB,[31][32] and many MATLAB libraries (for
example XML or SQL support) are implemented as wrappers around
Java or ActiveX libraries. Calling MATLAB from Java is more
complicated, but can be done with a MATLAB toolbox [33] which is
sold separately by MathWorks, or using an undocumented mechanism
called JMI (Java-to-MATLAB Interface),[34][35] (which should not be
confused with the unrelated Java Metadata Interface that is also called
JMI). Official MATLAB API for Java was added in 2016.[36]
As alternatives to the MuPAD based Symbolic Math Toolbox
available from MathWorks, MATLAB can be connected
to Maple or Mathematica.[37][38]

Libraries also exist to import and export MathML.

License:

MATLAB is a proprietary product of MathWorks, so users are subject

to vendor lock-in.[40][41] Although MATLAB Builder products can
deploy MATLAB functions as library files which can be used
with .NET[42] or Java[43] application building environment, future
development will still be tied to the MATLAB language.

Each toolbox is purchased separately. If an evaluation license is

requested, the MathWorks sales department requires detailed
information about the project for which MATLAB is to be evaluated.
If granted (which it often is), the evaluation license is valid for two to
four weeks. A student version of MATLAB is available as is a home-
use license for MATLAB, Simulink, and a subset of Mathwork's
Toolboxes at substantially reduced prices.

It has been reported that European Union (EU) competition regulators

are investigating whether MathWorks refused to sell licenses to a
competitor.[44] The regulators dropped the investigation after the
complainant withdrew its accusation and no evidence of wrongdoing
was found.
CONCLUSION AND FUTURE SCOPE:

Industrial product quality is an important part of product production,

and the research on defect-detection technology has great practical
significance to ensure product quality. This article provides a
comprehensive overview of the research status of product defect-
detection technology in complex industrial processes. Meanwhile,
combined with the actual application requirements and the
development of artificial intelligence technology, the defect-detection
equipment was investigated and analyzed. Through investigation, we
found that object detection, high precision, high positioning, rapid
detection, small targets, complex backgrounds, detection of occluded
objects, and object associations are the hotspots of academic and
industrial research. We believe that the investigation will help
industrial enterprises and researchers understand the research progress
of product defect-detection technology in the field of machine
learning and traditional defect detection. As future work, the proposed
approach for fabric defect detection should be optimized for speed
and its robustness tested with fabrics with different patterns using
deep learning techniques.
REFERENCE:

 [1] Fabric defect detection based on a deep convolutional

neural network using a two-stage strategy Xiang Jun ,
Jingan Wang, Jian Zhou, Shuo Meng, Ruru Pan and
Weidong Gao 2020
• [2] FABRIC DEFECT DETECTION USING DEEP
LEARNING K.S.Murugesan*1, S. Kiruthika*2, V.E.
Kowsika*3, M. Mohanapriyan*4, P.Naveen Kumar*5 2021

• [3] A Survey of Surface Defect Detection of Industrial Products

Based on A Small Number of Labeled Data Qifan Jin (1), Li
Chen 2019
• [4] Automatic Fabric Defect Detection Based on a
Convolutional Neural Network and Repeated Pattern Analysis
Yubo Huang and Zhong Xiang 2022
• [5] Computer Vision-based Fabric Defect Detection: A Survey
Ajay Kumar 2020

• [6] B. N. Nickolay and H. Schmalfub, ``Automatic fabric

inspectionUtopia or reality?'' Mellind Textilberichte, vol. 73, pp.
3337, Jan. 1993.
• [7] A. Kumar, ``Computer-vision-based fabric defect detection:
A survey,'‘ IEEE Trans. Ind. Electron., vol. 55, no. 1, pp.
348363, Jan. 2008.

• [8] E. J.Wood, ``Applying Fourier and associated transforms to

pattern characterization in textiles,'' Textile Res. J., vol. 60, no.
4, pp. 212220, Apr. 1990.

• [9] M. Unser and F. Ade, ``Feature extraction and decision

procedure for automated inspection of textured materials,''
Pattern Recognit. Lett., vol. 2, no. 3, pp. 185191, Mar. 1984.

• [10] A. Monadjemi, M. Mirmehdi, and B. Thomas,

``Restructured eigenlter matching for novelty detection in
random textures,'' in Proc. Brit. Mach. Vis. Conf., Sep. 2004, pp.
637646, doi: 10.5244/C.18.66.