Professional Documents
Culture Documents
Introduction to Artificial
Intelligence Jury Submission
Submited by:-
Mansi chand
Khushi gupta
Akriti pandey
Nisha kumari
ACKNOWLEDGEMENT
It is our proud privilege and duty to acknowledge the kind of
help and guidance received from several people in preparation
of this document. It would have never possible for me to make
this document without the kind of help and guidance I have
received.
First and foremost, I wish to record our sincere gratitude to Mr.
Abhay Kumar for his constant support and encouragement
throughout this whole semester.
The classes of introduction to artificial intelligence are really
helpful and it has contributed a lot in this documentation.
Last but not the least, we wish to thank our parents for
financing our studies in this college as well as for constantly
encouraging us to learn this course.
Mansi chand
Khushi gupta
Akriti pandey
Nisha kumari
Introduction to automation in spreading and cutting
By using an auto spreader, garment many manufacturers have reduced manpower in the
cutting room from 6 persons to 1-2 persons per cutting table (spreading activity). At the same
cutting room capacity is increased many times. Additionally, the most important thing, cutting
quality has been improved.
A lot of technology advancement happened in the spreading machine too at the IoT (Internet of
Things) era. Spreading statistic is available with the auto-spreader.
With the help of automation, a cutting room has become the most advanced department at an
apparel manufacturing enterprise. The cutting room management software is able to schedule
jobs to production, select optimal raw material and utilize remnants, generate cut plans and
nests, track production activities, and provide management reports for analysis and future
planning. Automated spreading machines significantly reduce the work load of the spreading
operator and increase work productivity. Automated marker making methods to match
material pattern significantly improve processing of garments from intricate pattern fabrics.
Fully automated knife cutting systems are the most effective means of cutting garments. They
ensure sufficiently high quality and accuracy, as well as, very high productivity for both low-ply
and high-ply cutting processes. By help of automation advanced fusing presses perform very
sensitive fusing process to ensure high-quality fused composite and avoid textile material
shrinkage.
Spreading
Fabric spreadingis a method where piles offabrics are spreadaspecific length and width wise
according to the garment markermeasurement.
1. Manual
2. Semi automatic
3. Fully automatic
1. Manual method:
Fabric is laid completely by hand. With the help of mechanical assistance as roll cutter. The
operator clamps the free end of fabric in line with the end of the spread. Pushes the spreader
to the other end. Cuts off the ply in line with that end. Clamps the beginning of the next ply.
Pushes the spreader to the other end other end and so on.
The spreading speed is dependent on the properties of a fabric. It is much faster (approx 1.5–3
times) than manual spreading. It is used in medium and large production enterprises, where the
volume of production justifies the cost of the equipment and its maintenance.
The same person checks the fabric edge, remove unnecessary creases on the fabric layer and
collect end bits.
The fabric spreader reaches to the end of the marker and starts laying the fabric on the table.
When the spreader reaches to other end of the marker, auto end cutter split the fabric from
roll.
Spreading table
Fabric feeder Spreading table
Spreading equipment
Cutting device Spreading machine
High productivity
Only one operator
Ability to spread all kind needed
of materials Low work load for an
Advantages of the method
Low expenses operator
Low productivity
Two operators needed High expenses
High work load for an Cannot spread materials
Disadvantages of the method
operator with intricate patterns
Fabric Cutting:
Cutting department plays very important role in any apparel industry. To make a complete
garment, cutting is necessary. Without cutting process to make a cloth is out of dream. Cutting
is the process which cut out the pattern pieces from specified fabric for making apparel with the
help of marker. Normally the marker is applied (drawn, traced, sprayed, clipped, pinned) on to
the top ply of a lay. Consistent fabric cutting is responsible for perfect garments making.
According to the working process, the cutting machine may be classified in three categories.
Such as-
Straight knife cutting machine is the most popular and versatile cutting machine. It is widely
used in clothing industry. Because it’s production speed is very high. Higher lay of height can be
cut very easily. Besides, knife is comparatively cheap and can be transferred from one place to
another easily.
Fig 3. straight knife cutting machine
Round knife cutting machine is called because it’s cutter is round in shape. It is also popular
cutting machine. This machine is small in size, flexible & used for small production. Round knife
is widely used for cutting the big parts of the garments pattern.
Computer numerical control may be consider to a means of operating a machine through the
use of discrete numerical values fed into the machine where the required 'input' technical
information is stored on a kind of input media such as floppy disk, hard disk, CD ROM, DVD, USB
flash drive, or RAM card etc. The machine follows a predetermined sequence of machining
operations at the predetermined speeds necessary to produce a work piece of the right shape
and size and thus according to completely predictable results.
Advantages of CNC
Easier to program
Easy storage of existing programs;
Easy to change a program;
Avoids human errors
CNC machines are safer
Complex geometry is produce as cheaply as simple ones
Usually generates closer tolerance than manual machines.
Disadvantages of CNC
1. Control systems
Open loop systems have no access to the real time data about the performance of the system
and therefore no immediate corrective action can be taken in case of system disturbance. This
system is normally applied only to the case where the output is almost constant and
predictable. Therefore, an open loop system is unlikely to be used to control machine tools
since the cutting force and loading of
a machine tool is never a constant. The only exception is the wire cut machine for which some
machine tool builders still prefer to use an open loop system because the is virtually no cutting
force in wirwcut machine.
Input output
System
In a close loop system, feedback devices closely monitor the output and any disturbance will be
corrected in the first instance. Therefore high system accuracy is achievable. This system is
more powerful than the open loop system and can be applied to the case where the output is
subjected to frequent change. Nowadays, almost all CNC machines usethis control system.
System
Feedback
I. Input Device
II. Machine Control Unit
III. Machine Tool
IV. Driving System
V. Feedback Devices
Introduction:
In this machine, cutting knife movement is controlled by a computer. Cutting table used for
computer controlled straight knife cutting machine is different from normal cutting table. The
table top is covered with a bed of nylon bristle. The nylon bristle bed is capable to support the
fabric lay and also allows the straight knife to pass through and move around the nylon bristle.
Hence cutting of each and every ply of fabric in the lay is confirmed. The nylon bristle bed also
allows air suction through the bed to make the lay compressed by vacuum pressure.
Working procedure
On the cutting table there are two rails on two sides, running length direction of the
cutting table. A cross beam moves length direction of the cutting table, situated on the
rails.
The cutting head is situated on the cross beam and moves on the cross beam but the
movement is in width direction of the cutting table.
The cutting knife is situated inside the cutting head. Due to the combined movement of
the cross beam, cutting head, and cutting knife, fabric cutting is done at any line, angle,
and point very accurately.
The combined movement is controlled by three servo motors.
When fabric cutting command is given through computer, the cutting knife starts lay
cutting from a reference point starting in the computer memory of the marker.
Notch point is made by the same cutting knife. Drill mark is also produced by a separate
drill situated in the cutting head.
Speed of cutting depends on lay height.
As the cutting proceeds, the vacuum pressure may drop into the lay.
To avoid the reduced vacuum pressure problem, another polythene sheet roll may be
used to cover the cut portion of the lay
Computerized straight knife cutting machine does not require any use of marker. In
some cases, marker may be used to identify the cut components.
Advantages:
Disadvantages:
A water jet cutter is an industrial tool capable of cutting a wide variety of materials using a very
high-pressure jet of water or a mixture of water. The cutter is commonly connected to a high-
pressure water pump where the water is then ejected from the nozzle, cutting through the
material by spraying it with the jet of high-speed water. The features of water jet cutting
machine is given below:
Fig 11. Waterjet cutter
Working procedure
1. In water jet cutter, water or mixture of water is used to cut fabric which is controlled by
computer.
2. Special software is used to cut fabric.
3. A very high velocity (60,000 lb/ square inch) & small diameter steam of water is created
by a water jet intensifier pump.
4. The high pressure jet acts as a solid tool & sharp knife which can easily cut the fabrics,
lathers & plastic materials.
5. As the jet penetrates successive plies in a spread, the momentum decreases & cutting
ability is reduced. So, the lower lays of fabric cutting will be wide & less effective.
6. The jet of water & loose fibers normally caught & drains away by a catcher when cut the
fabric. It is in the bottom of the fabric lays & moves with the same speed & the same
direction of water jet.
7. To improve the cutting speed, it is needed to adjust the pressure & radius of the jet.
Features
Plasma cutters work by sending an electric arc through a gas that is passing through a
constricted opening. The gas can be shop air, nitrogen, argon, oxygen. etc. This elevates the
temperature of the gas to the point that it enters a 4th state of matter. The restricted opening
(nozzle) the gas passes through causes it to squeeze by at a high speed, like air passing through
a venturi in a carburetor. This high speed gas cuts through the molten metal. The gas is also
directed around the perimeter of the cutting area to shield the cut.
In many of today's better plasma cutters, a pilot arc between the electrode and nozzle is used
to ionize the gas and initially generate the plasma prior to the arc transfer.
The advantages of robotic plasma cutting are numerous:
Cutting Speed
Wide Range of Material and
Ease of Use
Multi-tasking
Safety
Economical
Keeps Materials Cool
Artificial intelligence
Introduction
There is an increased need for quality control in the manufacturing sector. The apparel industry
rejects any materials with defects in manufacturing processes because a minor defect in a
manufactured part might result in a disaster at a later stage. Without automatic machine vision
technology, cutting operations is not able to perform real-time inline cutting defect inspection.
The failure to identify some defects can lead to factory down time and significant economic
losses.
The many factors make defect features difficult to detect: material impurities, fluctuation of
lighting, object twitter and other unexpected movement between them fabric and the cutting
machine.
Our suggestions efforts have been put in modeling defect detection as a pattern classification
problem. There are two important issues to be addressed.
First, effective defect features must be explored, extracted and optimized.Defect features
characterize cutting defects and determine the complexity of the classification system.
The second important issue is the classifier design, which determines the performance of
the entire vision system in classifying defects.
We have developed a machine learning system based on the Support Vector Machine. During
the design of the machine vision system, we address three main requirements:
1. It should be robust and capable of providing good discrimination even in the case of noisy
input data.
2. It must be fast enough in order to meet inline speed requirements.
3. It should be incrementally scalable to incorporate new known objects, particularly new
known defect classes, without retraining with the whole data set. In addition, we address
the issues including parameter selection in SVM.
The system has been implemented and tested on image data downloaded directly from fabric
Cutting productions. The experiments showed that the machine vision system meets the real-
time requirement with a highdefect detection rate.Features characterize surface defects and
determine the complexity of the classification system.
Figure1. A machine vision system for defect detection.The rough filtering algorithm explores
several defect features including the orientation of the grain line, which is approximately
vertical in the images. The rough filtering algorithm applies a horizontal gradient operator to
the image to detect the edges of the fabric. If a filtered image has no significant gradient values,
the image is considered normal. However the horizontal gradient filtering also generates false
seams
The defect candidate finding algorithm explores fact of the edge, and usually has high contrast
with theedge being always darker than its surrounding area. This implies that a horizontal
gradient filter usually produces significant negative values on the left side and significant
positive values on the right side of a edge.
The algorithm first searches for the initial edge points, called seeds. A seed point is the local
maxima in the gradient image and has either negative-to- positive or positive-to-negative
neighbors on the two sides. If the minimum edge length is set to be n pixels long, then the
gradient image is sampled every n rows to search for seeds. The second step is to grow the
seam region starting from the seed points. It begins by examining six- and the candidate region
is identified..
Cut order planning is one of the most complex processes involved in garment making. AI can
make the planning efficient and faster. For example JAVA software has developed an AI
integrated cut planning software- OPTACUT which can initiate fabric required for ordering
almost accurately and makes the cut planning process automated.
The garment manufacturing process is becoming more automated to cater the increasing
demand of consumers, reduce the number of faults, and keep the production cost low.
However, the industry still lacks the extensive adoption of AI methods. The industry is still using
computational tools based on classical algorithms and modern AI techniques are confined to
academic research. Hence, it is a requisite for the industry to adopt new AI techniques to have a
competitive advantage and improve business profitability.
One of the most important aspects where AI can be used is automated workflows based on
predictive analytics.
With the help of AI, detection and solving a problem becomes easier. They are programmed to
detect the faults and fed with solutions to overcome the damage.
They are making production decisions smarter and instant. With the help of AI, labor cost can
be minimized.
Advises for application of artificial intelligence using machine learning and
computer vision technologies
Challenges addressed-
Prevalent spreading processes can require significant investment of time and knowledge to
determine the spreading mode according to the fabric type and also specifying technical
information like ply tension, no. of plies etc that are required for spreading by an automatic
machine, that will be adequate to meet client specifications.
Secondly, the operator that will be operating the spreading machines needs to have technical
knowledge to work on CNC systems of the automatic spreading machines.
Opportunity
AI can be used to find patterns in spreading data that can lead to the best possible spreading
solutions. By looking at a wide variety of data including fabric properties, spreading modes for
fabric types , prior configurations like tension of ply , no. of plies etc .
AI based models can determine which combinations of variables are most likely to produce a
positive result. Using this information, garment technicians can pursue avenues that are most
likely to work and may even find new solutions they had not thought of before.
By focusing on high-probability solutions, manufacturers can reduce costs, speed time to
market and improve quality.
Reference for the idea is from H2O it is an open source machine learning platform.
(site: https://www.h2o.ai/)
Where ,
Machine Learning and Computer Vision cogniti ve technologies that emanate from
AI can be used in the applicati on of arti fi cial intelligence in Spreading equipments.
Computer vision (CV) is a process (and a branch of AI ) that involves capturing, processing and
analyzing real-world images and video to allow machines to extract meaningful, contextual
information from the physical world.
Machine learning is the process of teaching a computer system how to make accurate
predictions when fed data.
Machine-learning algorithms use statistics to find patterns. The data is digitally stored, and it
can be fed into a machine-learning algorithm.
Spreading Modes- Two fabric characteristics that determine the spreading mode are the
direction of fabric face and the direction of the fabric nap.
Other information will be provided to the system like Garment style, garment quantity,
garment sizes etc
Using these data and information it will suggest the most optimum way of spreading the
plies.
The system will then automatically spread the fabric in the correct spreading mode ,No. of
plies according to garment quantity and in accurate tension according to the fabric type.
Fig; Set-up
Exampleof of
theImages
CCD camera forare
like that fabric analysis
acquired forusing Computer
scanning of thevision
fabrictechnique
Figure: Microscopic image of fibres to identify fabric content
Fig: scan image of woven fabric Fig: scan image of rib knitted fabric
And thereby providing the CNC systems of spreading machines the ability to automatically learn
and improve from experience without being explicitly programmed.
These are the leading brands in the garment industry that provide automatic spreading
equipments.
1. Lectra: Lectra is a well-known brand in India and worldwide for cutting room automation in
the sewn product industry. Lectra spreading machine is very popular in India.
2. Gerber: When we talk about cutting room automation, Gerber comes after Lectra. Gerber
and Lectra spreaders are found in most of the garment factories in India. Know more
about Gerber spreader.
3. Eastman: Eastman company is synonyms of the straight knife cutting machine. They have
recently introduced cutting room automation including Eastman automatic spreading machine.
9. IMA Fabric spreaders: IMA SpA is an Italian capital company founded in 1977 in Turin. A few
years after its establishment, it joined the Macpi Group and transfers its production site in
Palazzolo, in the province of Brescia, home of the other companies of the Group.
10. Ozbilim: Ozbilim is a Turkish company. Recently I visited Turkey. I have seen Ozbilim
spreading machines in a garment factory.
11. YIN (Automatic Spreader) // Yin USA, Inc. provides integrated automation solutions for
advanced textile manufacturing industries by including Automatic Spreading Machines.
Conclusions
This represented a defect detection approach based on machine vision and intelligent systems.
The approach fully explored visual defect features, and developed a SVM learning algorithm
that automatically learns complex decision boundaries when data noise exists. Experimental
results demonstrate the potential of a Support Vector Machine as a promising classifier for
defect data with application in real-time manufacturing environment. The in-line tests are
currently underway. We expect that the system will help the garment industry to improve the
consistency of product quality and production
REFERENCES
https://www.h2o.ai/manufacturing/?
gclid=EAIaIQobChMI9fHn4dis6gIVRiQrCh38bwK_EAMYASAAEgKg4fD_BwE
https://www.h2o.ai/solutions/usecases/predictive-manufacturing-design/#solutions
https://medium.com/@miccowang/computer-vision-the-closet-thing-to-ai-on-our-personal-
device-d2ff63994856
https://jaza-soft.com/blogs/application-of-artificial-intelligence-in-the-apparel-industry-2
file:///C:/Users/khushi/Downloads/AIintextiles_ATJ_manuscript.pdf
https://valiancesolutions.com/applications-artificial-intelligence-ai-textile-industry/