The document is a project report submitted by Mr. Gaurav U. Dandekar as part of his Bachelor's degree in Textile Science at Nagpur University, focusing on fabric defects and their automated detection methods. It highlights the importance of quality control in the textile industry, detailing various types of defects, their causes, and solutions, as well as the benefits of using machine vision for improved inspection accuracy. The report concludes that minimizing fabric defects is crucial for maintaining product quality and reducing costs in textile manufacturing.

1. Created By – Rizwan Rajik Qureshi

2. Project Report
On
Submitted by
Mr. Gaurav U. Dandekar
(Bachelor of Textile Science, III Year)
Under the Guidance of
Mrs. Snehal Rohadkar
BACHELOR OF TEXTILE SCIENCE
MAHALAXMI JAGDAMBA COLLAGE OF LIBRARY & INFORMATION SCEINCE
RASHTRASANT TUKDOJI MAHARAJ NAGPUR UNVIVERSITY,
NAGPUR.2018-2019
File
notcopy

3. Is a record of dissertation work
Carried out by
Mr. Gaurav U. Dandekar
Submitted in the partial fulfilment of requirement
For the degree of Bachelor of Textile Science
Of R.T.M. Nagpur university.
MAHALAXMI JAGDAMBA COLLAGE OF LIBRARY & INFORMATION SCEINCE
RASHTRASANT TUKDOJI MAHARAJ NAGPUR UNVIVERSITY,
NAGPUR.2018-2019
Mrs. Snehal Rohadkar
Guide
Mrs. Janvi Nandanvar
Principal
Mrs. Meghna Polkat
Head of
Department
File
notcopy

4. Acknowledgements
It was golden opportunity for these project and each step is a learning
process of one life. Each opportunity that we get adds something to our
personality. I take this opportunity to express my profound gratitude
towards every who helped me through the making of this project.
To being with I must acknowledge the whole hearted support I received
from Mrs. Snehal Rohadkar and Mrs. Meghna Polkat.
I am also thankful to Principle Anita Tekade Madam for their valuable
guidance. Mahalaxmi Jagdamba Mahavidhylaya Nagpur.
I would like to express thanks to Mr. Rupesh (Employe in Shurya lakshami
ltd.) My outmost thanks to all the staff members for their valuable
support, encouragement, guidance and suggestion during the Project
Finally I thank to all my friends and each and every member for their
unfailing, valuable suggestion, support & assistance.
Mr. Gaurav U. Dandekar
B.T.S Final Year
Mahalaxmi Jagdamba Mahavidhylaya
Nagpur
File
notcopy

5. Contents
01 - 10
01 - 10
01 - 10
01 - 10
01 - 10
01 - 10
05
CONCLUSION
06
REFERENCE
01
INRODUCTION
1. FABRIC DEFECTS
2. WEAVING DEFECTS
3. FABRIC DEFECT DETECTION
SYSTEM
4. GENERIC FABRIC DEFECTS
DEFECTION SYSTEM OVERVIEW
02
STUDY OF FABRIC DEFECTS
03
FABRIC FAULT IN WEAVING
04
DESCRIPTION OF FABRIC DEFECTS
1. WEAVING
2. DEFINITION OF DEFECTS
3. FABRIC DEFECTS
4. FABRIC INSPECTION
SYSTEM
01 - 04
05 - 06
07 - 09
10 - 26
27 - 28
29 - 29
File
notcopy

6. 1
I N T R O D U C T I O N
File
notcopy

7. 2
 Introduction
Quality control is an important factor in industrial production which is increasing
day by day. Textile industry is not an exception in this regard. Inspection of fabric defects
plays an important role in the quality control. Human vision inspection is not enough
accurate due to fatigue and tediousness. Also, it is time consuming and high quality
cannot be maintained. So the solution to this problem is automated i.e. machine vision
based textile inspection system. Automated textile inspection involves two challenging
problem, namely, defect detection and defect classification. Classification process is
composed of several steps. One of the important steps is scene analysis and feature
selection. The complexity of the subsequent steps increases and the classification task
becomes hard by selecting an inappropriate feature set. Bangladesh, as a developing
country specially focuses on export, through which a lot of foreign exchange is earned.
Bangladesh textile industry is major export sector and strength of industry are cheap labor
cost, low energy price, and good-quality products. In the highly competitive global market
to increase the performance industry should increase productivity as well as quality.
Quality is affected due to defects that occur in textile. So accurate detection of defect is
an important factor to improve the quality. Defect analysis is a very important part of our
approach to automated textile defect inspection problem, which has been done earlier
than all other parts. Defect analysis helps understand the defects properly, and give clues
to appropriate feature. In this paper, we have dealt with four types of defects, which
frequently occur in knitted fabrics in Bangladesh, namely, color yarn, hole, missing yarn,
and spot. Automated textile inspection systems are real-time applications. So they require
real-time computation, which exceeds the capability of traditional computing. Neural
networks (NNs) are suitable enough for real-time systems because of their parallel-
processing capability. Moreover, NNs have strong capability to handle classification
problems. The classification accuracy of an appropriate NN, which handles multiclass
problems, is good enough. There is a number of performance metrics of NN models.
Classification accuracy, model complexity and training time are three of the most
important performance metrics of NN models. Objective of this paper is to present an
appropriate set of features in order to address the problem of defect classification. We
justify the features in terms of their discriminatory qualities considering the difficulties
which lie in the feature extraction process. We use statistical approach to extract the
features.
1
File
notcopy

8. 3
FABRIC DEFECTS
Defect is an area of fabric that does not conform the features of fabric or fulfill
the requirements which increase the dissatisfaction of customer. There are three main
categories of defects as follows
Captious defects- Injurious for health of the user.
Major defects- Highly affect the acquisition of product.
Insignificant defects- These do not effect on purchasing of product.
Weaving defects
These kind of defect occurs when thread has interlocking process. Broken ends,
float, broken pattern, holes, gout, cut, oil stain, grease spot, water stain, heap and mille
defect & missing end come under weaving defects. The textile industry, as with any
industry today is very concerned with quality. It is desirable to produce the highest quality
goods in lower cost and shortest amount of time possible. Fabric defects detection is an
important part of quality control in the textile industry. In this paper we analyze the faults
using image processing techniques. Image processing techniques will help to increase
the production of fabric industry as well as increase the quality of products.
Fabric Defect Detection System
Detection is a process of inspection of fabric whether cloth has fulfilled all the
related specifications or not. There are two types of inspection system are used for defect
examination.
Manual Defect Detection
Manual inspection is done by the humans. For this, textile industry is needed a
very highly trained and experienced inspectors. A very extent ratio of fault has being
identified upon cloth material in textile Manufacturing Corporation. It is costly and time
consuming. Highly trained inspector can detect only 70% of defects.
File
notcopy

9. 4
Automatic or Semi-automated Defect Detection
Automatic defect detection is done by the machine; therefore inspection
based on automatic system can increment the ratio of accuracy in a high order. It is a
real time based detection method. Using this method, textile industry having a good and
100% inspected fabric. It is more reliable, fast inspection; reduce labor cost and
improvement in fabric texture.
GENERIC FABRIC DEFECT DEFECTION SYSTEM
OVERVIEW
The ultimate goal of Fabric defect detection system is to be able to identify the
defects and it’s type no matter what the circumstances (shadowing, lighting effects etc.).
As this is not trivial to achieve, a step by step approach, starting from segmentation to
defect identification has to be developed. The method for the whole system can be
characterized by the series of steps. The flow graph of generic fabric defect detection
system consisting of basic steps is shown in following ( Fig. ).
Fabric image acquisition
|
Preprocessing
|
Filtering
|
Threshloding and segmentation
|
Feature extraction
|
Classification of defect
|
Recognition
Fig. General Defect detection flow chart
Firstly the fabric image is capture with digital camera. After that preprocessing approach are
applied to that image (like noise removal, histogram equalization etc). Further, filtering operation is
performed. With the next proceeding Threshold using segmentation are applied. So that image
feature is extracted for neural network and fuzzy logic (for humanistic approach). To complete the
detection apply neural network and fuzzy logic in order to detect and classification of defects.
File
notcopy

10. 5
S T U D Y O F
FA B R I C
D E F E C T S
File
notcopy

11. 6
 Study of Fabric Defects
 Due to the increasing demand for quality fabric, high
quality requirements are today greater since customer
has become more aware of poor quality problems
 To avoid Rejection of fabric, it is necessary to avoid
defects.
 Price of fabric is reduced by 45% - 65% due to the
presence of defects.
 Company image will go down.
2
File
notcopy

12. 7
FA B R I C FA U LT
I N W E AV I N G
File
notcopy

13. 8
 Fabric Fault in Weaving
Weaving:
Weaving is the intersection of two sets of straight yarns, warp and weft, which cross
and interlace at right angles to each other. The lengthwise yarns are known as warp
yarns and width wise yarns are known as weft or filling yarns and the fabric produced
is known as woven fabric. The machine used for weaving fabric is a loom. It is a
complex work. A number of faults occur in fabric during weaving process.
Definition of Defect:
An imperfection that impairs worth or utility.
Or, Want or absence of something necessary for completeness or perfection.
Or, A fault that spoils the material.
Fabric Defect:
A Fabric Defect is any abnormality in the Fabric that hinders its acceptability by the
consumer. 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
the production speeds make the timely detection of fabric defects more important
than ever.
Fabric inspection system:
Definition of inspection:
Activities such as Measuring, Examining, Testing, one or more characteristics of a
product or source comparing these with specific requirements to determine
conformity. The fabric roll is removed from loom and then sent to an inspection
frame.
Object of fabric inspection:
The word of Inspection in textile engineering mainly identifies to the visual
examination or review/scrutinize of materials (like fabric, accessories, trims etc). It is
an imperative process in any woven industry to keep away from rejects due to low
fabric quality and precautions to remove unexpected loss/defects of finished goods.
The quality of a finished item in woven industry mostly be controlled on the quality of
fabric when it is produced as a roll from. Even it occupies 60/70 percent of total cost.
Apparel industry should take some defensive actions to ensure quality full fabric is
being used in their items and set up fabric inspection department with modern
3
File
notcopy

14. 9
equipment and skilled manpower. 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.
Fabric mills and garments industry, both parties quality team should give their
highest effort to erase any unexpected circumstances. Even the most
exceptional/excellence manufacturing methods cannot compensate for defective
materials.
Fabric faults in weaving
Example of some samples.
File
notcopy

15. 10
D E S C R I P T I O N
O F FA B R I C
D E F E C T S
File
notcopy

16. 11
 Description of Fabric Defects
Weft bar
It is a band running weft-wise across the full width of the cloth. The normal reasons
are the periodic medium to long term irregularity in the weft yarn, count difference in
weft, excessive tension in the weft feed package, especially in filaments, variability in
pick density and difference in twist, colour or shade of adjacent group of picks,
difference in blend composition or in the cottons used.
Causes:
If count of yarn vary from cone to cone or within the cone then a bar of weft will
appeared in fabric after weaving.
Remedy:
To solve this problem we need to buy yarn with uniform thickness.
4
File
notcopy

17. 12
Thick and thin places
These are similar to weft bar, but unlike weft bars, it repeats at intervals.
They are mainly due to irregular let-off, incorrect setting of holding and releasing
pawls on the ratchet wheel of take-up motion, gears of take-up motion not meshing
properly, and gear wheel teeth worn out or broken.
Causes:
This can happen due to not removing of spinners double, not piecing the end
properly by removing the lapped materials, accumulation of fluff in condensers,
cradles and in the necks of the top rollers.
Remedy:
The yarn have to be checked by QC during the time of procurement to avoid these
faults
File
notcopy

18. 13
Bad selvedge
Causes:
picks yarn the bad selvedges is
occurred.
the temple is not properly used during weaving.
the Heald Eye and Heald shaft.
If it does not complete carefully the selvedges will be poor.
endar is cut by the press roller.
If the stanter machine stretches the fabric more.
Remedies:
carefully.
File
notcopy

19. 14
Broken picks
A filling yarn that is broken in the weaving of a fabric appears as a defect.
Improper functioning of weft stop motion results in broken picks undetected and
going in to the fabric.
Causes:
the beat up motion is done before insertion of full pick, this type of fault may be
occurred.
Remedies:
File
notcopy

20. 15
Broken end
A defect in fabric caused by a warp yarn that was broken during weaving or finishing.
Causes:
e dropper bar , then this
fault arise.
Remedies:
observe this problem.
File
notcopy

21. 16
Double end
Two ends that weave as one. This happens because of migration of a broken end to
the adjacent reed space along with the neighbouring end.
Cause:
that properly than those ends can go together through the same heald eye and can
create this problem.
Remedies:
y carefully that, two or more ends cannot be
knotted with each other.
.
File
notcopy

22. 17
Knots
Knot is defined as a knob or lump formed by interlacing portions of one or more
flexible strands or a quantity of yarn, or thread, which varies with the fibre; it consists
of a set of coils. Control in pirn winding, the winding to binding coils ratio can solve
this problem.
Causes:
the yarn breakage during winding, warping, sizing and weaving this fault may be
occurred.
Remedies:
The extra foreign matter can be pulled out with a plucker. Combing in both direction
rectifies the resultant patch.
File
notcopy

23. 18
Miss pick
A defect in woven fabric caused by a missing or out-of sequence yarn
Cause:
If a worker start a stopped machine without picking the broken weft from the shed
then this type of fault arise.
Remedy:
We have to motivate the worker to do job properly during time of pick finding.
File
notcopy

24. 19
Reed mark
A crack between groups of warp ends, either continuous or at intervals, which can
happen due to damaged reed or improper spacing of dents.
Causes:
of warp ends, either continuous or at intervals, which can
happen due to improper spacing of dents.
Remedy:
The reed have to be changed .
File
notcopy

25. 20
Tight end
Warp yarn in a woven fabric that was under excessive tension during weaving or
shrank more than the normal amount.
Causes:
s present in the loom then
this problem arise.
over tension produced on warp during warping and weaving.
Remedy:
To solve this problem we have to find these ends and have to adjust the tension
File
notcopy

26. 21
Crease mark
Causes:
roller
Remedies:
- creasing agents are used to avoid crease mark problem.
aintenance of the machine
File
notcopy

27. 22
Double ends
Two ends that weave as one. This happens because of migration of a broken end to
the adjacent reed space along with the neighbouring end.
Cause:
ds of warp sticks together. And if the worker don’t observe
that properly than those ends can go together through the same heald eye and can
create this problem.
Remedies:
or more ends can not be
knotted with each other.
File
notcopy

28. 23
Double pick
Causes:
may be happens because of migration of a broken end to the adjacent reed
space along with the neighbouring end.
Remedies:
File
notcopy

29. 24
Loose Pick
Causes:
happens due to the malfunctioning of the programmable felling tensioner.
-winder to fixed nozzle.
Remedies:
pressure of the main valve, relay valve and relay nozzle should be perfect.
-winder should uniform and straight.
File
notcopy

30. 25
Ball
Causes:
If the warp is too much hairy. Then the reed will create ball in warp yarn in- between
heald shaft and reed. If the ball is small enough to pass through the dent of reed
then they will form the ball in fabric.
Remedy:
To solve this problem we have to change the reed, or reduce the tension of the wrap
yarn.
File
notcopy

31. 26
Missing end
One warp end is missing in the fabric. Void caused by a missing warp thread in the
fabric.
Causes
 Failure of weaver in attending to warp breaks immediately.
 Warp stop motion not acting properly.
Remedies
 Missing the incidence of lappers during sizing.
 Use spare ends on loom as a substitute for the missing ends.
 Instruct the weaver to attend to warp breaks immediately.
 Discourage the weaver’s habit of waiting for the broken end on the beam to advance
sufficiently for knotting.
 Inspect the drop pins while putting on a new beam and weed out the defective ones.
 Check the warp stop motion assembly
File
notcopy

32. 27
C O N C L U S I O N
File
notcopy

33. 28
 CONCLUSION
Though majority of the fabric faults occurred in the weaving section, all the
other processes are also very important to get the excellent quality faultless
fabric.
Due to presence of fabric defect, it has to be sold at lower prices, or even in
some cases as seconds, which creates a Hugh value loss to the company.
To minimize value loss due to variety of defect occurring in the fabric, a
manufacturer should try to minimize those defects by taking suitable remedies.
An automated defect detection and identification system can enhances the
product quality and results in improved productivity to meet both customer
demands and to reduce the costs associated with off quality
I have done project work is very careful with successfully. I found in the
industry every person is very helpful and positive attitude. I also found the
fabric faults with its remedies in every division. The project is very essential in
our job life.
5
File
notcopy

34. Reference
http://data.conferenceworld.in/ESHM5/P337-347.pdf
http://textilelearner.blogspot.in/2013/07/weaving-faults-faultsdefectsp
roblems.html
http://textilelearner.blogspot.in/2013/07/defects-in-woven-fabric-with
-image.htm
https://textileapex.blogspot.in/2015/03/woven-fabric-defects-causes-r
emedies.html
http://shodhganga.inflibnet.ac.in/bitstream/10603/40774/17/17_appe
ndix.pdf
http://cbse.nic.in/publications/vocational/Textile%20Design/Woven%2
0Textiles.pdf
File
notcopy