Print this article

International Journal of Mechanical Engineering and Computer Applications, Vol 1, Issue 4,
July- August 2013, ISSN 2320-6349
Textile Image Processing for Real time Application
Roopa M R and Badari Nath K
Roopa M R Lecturer, PESPT, Bangalore E-mail:[email protected]
Badari Nath K, Asst .Professor, RVCE, Bangalore E-mail: [email protected]
ABSTRACT-- Electronics did not exist in the early part of
19th century, thus internal operation of Babbage analytical
engine and Jacquard’s loom were required to be mechanical.
Jacquard is a technology adopted among handloom and power
looms to produce required design on the woven fabrics like
sarees, dress materials and furnishings. Existing jacquard is of
mechanical, which requires laced punched card for design
feeding. By the use of instructions punched in a card, Jacquard
looms were capable of weaving intricate patterns, flowers,
leaves etc. This method is used in different parts of the
country from decades.
The work proposes the restoration of textile design from old
set of punched cards, available in different variety, sizes and
color. This helps to build digital design archive which
eliminates storing punched cards. Digitization of cards is
necessary as the cards may get damaged with repeated use,
environmental condition, missing of particular card while
storing etc. Since majority of textile industry in India use
design weaving through punched cards, digitization leads to
re-creation of new varieties and designs makes value added
addition, increased sales and production and also saves lot of
time. If we do not restore we may loose lot of traditional
designs developed by textile artists. Also Old digitized textile
designs can be easily re-punched using computerized
punching machines as and when required at low cost, high
speed and more accurately.
Images are fed to the computer using flatbed A4 scanner with
resolution of 300dpi. Processing the punched card image and
interpreting information in the card in terms of zeros and ones,
are the main objectives of the project. Processing of image is
done using MatLab7 software and binary output is saved in
bmp file. Bmp file is accessed by software and maps binary
coded information into design pattern. This also provides
facility to view design and digitized punched cards, also
facilitates editing of saved bit pattern.
Key words- textile design, analytical Engine, Image
processing.
INTRODUCTION
www.ijmca.org
One of the oldest method of producing cloth
manually is handloom. The word ‘handloom’ means
traditional method of weaving which employs simply a loom
operated by hand and was the basic activity of human society
since time immemorial in which utility and aesthetics are
blended together. The handloom industry is one of the cottage
industry having prevalent as well as primeval in India’s
decentralized sector. Saree is the traditional Indian women
wear and has been the ultimate hallmark of Ethic feminity
through the ages.
A ‘power loom’ is one more type of loom, which is
mechanized tool, uses a driven shaft for power. Power loom is
invented by Edmund Catwright in 1784. This loom allowed
textile manufacturer to produce more quicker than any handdriven looms. Many industries uses machines used for
weaving textiles which has sensors and alerts personnel when
yarns break or shuttle mechanism go awry. Each type of
power loom may be equipped with different operation and setups, each varying in different amount of human interaction.
A ‘Jacquard loom’ is a mechanical loom. It is
invented in 1801 by Joseph Marie. The manufacturing
complex textile patterns (brocade, damask and matelasse) is
simplified by using this loom. Punched card with punched
holes controls this loom. Each card represents one row of the
design.
Numerical and Character data were stored on
punched cards, where a complete record of several hundred
characters might span several cards. The data processing, or
other department, had to dedicate lots of cubic feet to the
storage of racks and racks of punched cards. Data processing
operators handled the punched cards and many record keeping
& reporting tasks were accomplished more or less
automatically. Central to a data processing operation were the
Card Sorters that were able to quickly & accurately drop cards
read from a stack into slots according to the holes punched in
a column. Gang punches, duplicating punches, and tabulators
handled record keeping and reporting tasks.
Textile Punched cards
The loom controlled by a "chain of cards", a number
of punched cards, laced together into a continuous sequence.
Multiple rows of holes were punched on each card and each
Page 70
International Journal of Mechanical Engineering and Computer Applications, Vol 1, Issue 4,
July- August 2013, ISSN 2320-6349
row of punched holes corresponded to one row of the design.
Several such paper cards, generally white in color, can be
found. Chains, like the much later paper tape, allowed
sequences of any length to be constructed, not limited by the
size of a card. The Jacquard loom was the first machine to use
punched cards to control a sequence of operations. Textile
punched card comes in variety of forms such as 256 size, 600
size depending upon design pattern. Design pattern depends
upon butta, border etc. Cards color is also not standard. Figure
1 shows textile punched cards.
changing the card, this ability was an important conceptual
precursor of computer programming towards development.
1.2 Digitization of patterns
Development of product
on
saree
and
Research includes following sections.
Selection of Motifs
The conventional motifs of embroidery consists of
animal, bird and many geometrical patterns. These motifs
may be categorized into Buttas , Borders etc design patterns.
These patterns are digitized and prepare punched cards for
jacquard .Examples of motifs are shown in figure2
b) Digitizing the conventional motifs
GC Kala -2004 with interface Paint Shop Pro
software was used to automate the selected conventional
motifs into woven designs. By simple handloom with
jacquard shedding mechanism is employed to weave the
motifs. Later it was checked with resemblance of automated
designs with hand embroidery motifs.
a)
Figure 2 Conventional Motifs
1.3
Figure 1 Textile punched card
MatLab
MatLab is a dynamic scientific programming
language that is commonly used by scientists because of its
convenient and high-level syntax for arrays, the fact that type
declarations are not required, and the availability of a rich set
of application libraries.
Making Jacquard pattern
As discussed GC punch software assists and controls
Jacquard card punching process. The software maintains order
of card to be punched too. According to design cards are
numbered serially and laced. Figure3 shows textile punched
card.
1.1 History of computing
Using punched card the Jacquard loom controlled a
sequence of operation. This was the first machine to use this
concept. Even though there involved no computation, it is
considered as an important revolution in the history of
hardware computing. Weave the change of pattern by simply
www.ijmca.org
Figure 3 Textile Punched card
Page 71
International Journal of Mechanical Engineering and Computer Applications, Vol 1, Issue 4,
July- August 2013, ISSN 2320-6349
1.4 Proposed System
Depending upon type of design there will be different sized
textile punched cards. For example 256size,600sized etc.
Irrespective of type proposed system converts the data in the
card to 0’s and 1’s. Region of punch is identified as 1 and
non-punch as 0.The software operates in real time. The system
tasks are as follows.
Select and Identify type of card
Read the file containing the cards through
scanning.
Store the binary data in the file.
Convert stored binary data into design.




I.
PRODUCT PERSPECTIVE
The purpose of this software is to read the punched
card data. This helps in restoration of the textile punched card.
The intended audience for this project are loom owners
(restoration of old punched card), designers (design
verification). This product helps textile loom owners to store
the punched card data and reading the data from the card
within fraction of seconds. If the quality of the card is low,
may loose the data and again card need to be punched. With
data stored in a system it becomes easy to punch the card.
Hence it minimizes hardware cost and card implementation
time. Manual entry in achieving data storage is prevented. The
option is also provided which edit the stored bit patterns in
case bit error or small change in design pattern.
Reading Image from memory Enough memory must be
provided to store the image in the system. Image can be stored
in JPG, Bmp etc format which does not affect our conversion
process.
Scanner A4 sized scanner is perfect to work with for smaller
and medium sized cards , but carefully keeping cards of
bigger size for A3 sized scanners or scan twice.
2.2
Functional Description
Textile Punched card processing system contains
various processing tasks to digitize the data. Non-technical
person should be able to operate and the software is the main
aim. Hence provided with number of buttons, through GUI
which enable loom worker to convert data from card to binary
format. Table1 shows the GUI.
Table 1 : Table of Functions
Function Name
Description
Browse_button_click
Allows to browse the card to be
converted
Convert_button_click
Converts selected card to
binary format and saves in
specified file
Show_button_click
Shows stored binary data to
design format
Clear_button _click
Clears for fresh card selection
Card_Select_Button
Allows Smaller or Bigger
Sized Card(256 or 600)
Edit_button_click
Allows user to edit stored bit
pattern
2.1 Functional Requirements
Ensuring design security The weave design is the result of
artist’s creativity and intellectual property. Hence it has to be
secure through all stages of human intervention. The format
has to be designed to take this aspect. Also option has to be
there to operate software by authorized persons, for example
loom owner.
User friendliness for weavers The weavers are not much
literate and they need intuitive user interfaces for operating
the computer.
Operating with computer Even with not much knowledge
about the software when employee wants to restore the image
pattern it should be provided with click button and access the
stored punched card image as input. Minimum input during
runtime must be asked such as number of columns. This
information is needed as the cards are not standard.
Computers have assisted to carry on the work easily and in a
short duration, maximum number of assignments can be
fulfilled leading towards profit of the company.
www.ijmca.org
II.
IMPLEMENTATION
In developing any application, the programming language
plays a significant role. The choice of the programming
language has to be made based on the requirements of the
application at hand. In this application, MATLAB
Page 72
International Journal of Mechanical Engineering and Computer Applications, Vol 1, Issue 4,
July- August 2013, ISSN 2320-6349
programming is used as it is very convenient to develop image
processing applications with it, owing to its wide variety of
available toolkits and libraries. However, only basic library
functions have been used for image processing toolbox. The
implementation of algorithms has been done stepwise.
The platform selected for the project is Windows 7, as it
is a compatible OS for running MATLAB. Also, the input
image for the application is in .tiff ,jpg ,gif etc. which is a
multimedia container format introduced by Microsoft in
November 1992 as part of its Video for Windows technology.
These files are compatible both with Windows 7 and
MATLAB.
Step 10: Find distance between top2 and bottom2 lacing holes,
divide by 5, gets inter pixel distance. (Distance x distance y).
Step 11: From first top punched region (lacing hole) move 3
distance down, gets the 2nd punched region, move one distance
(interpixel) left, gets 1st punched region.
Step 12: From 1st punched region travel across all regions (8
columns and 32 rows) along the distance calculate mean pixel
value .
Step 13: Interpret value 0 or 1 along each centroid depending
upon mean pixel value.
III.
3.2 Implementation Method
Image of textile punched card can be obtained by any
means such as camera, scanner etc. Scanner with resolution
300dpi is preferred as it is an accurate method, easy to operate
such as scanning card is friendly and we get clear reference at
the edge, low bit error, algorithm works fine even if there is
slight misalignment of card inside scanner. However card
need to be scanned twice if it is bigger size (600).
RESULTS AND ANALYSIS
Textile Image Processing for Real-Time application
mainly deals with reading of information present on Textile
Punched Card. Process images of punched cards by using
software in a computer and read information present in a card.
Interpretation of bit pattern on a card, which is a kind of
reverse process. Set of cards laced together to give predefined
design pattern. These are patterns which is woven on a saree.
Below steps explains how matrix of binary is obtained by
reading punched card.
Step 1: Load image and read using matlab command.
Step 2: Initialize empty array ,Number of cards to be inputted
and Min –Max area.
Step 3: Convert Image to black and white if it is color using
thresholding function.
Step 4: Highlight background so that punched regions are
clearly visible(contrast by subtracting image from 1 if it is
white and 0 if it is black).
Step 5: Remove all object with pixel value less than 280
Step 6: Using BWlabel recognize contours and extract their
properties of type bounding box using regionprops.
Step 7: Fill an array of matrix with co-ordinate value of
recognized contours with area within Min-Max threshold area.
Step 8: Array contains x, y co-ordinates of contours. Find the
centroids.
Figure 4 Design obtained from 20 cards
Step 9 : Sort x and y co-ordinates according to their position.
www.ijmca.org
Page 73
International Journal of Mechanical Engineering and Computer Applications, Vol 1, Issue 4,
July- August 2013, ISSN 2320-6349
Figure 4 shows snapshot of selecting cards for inputting and
obtained binary and design pattern at output.
IV.
SUMMARY
Implementation of punched card is made generic
irrespective of type of cards. Implementation took considering
lacing holes as references which are present in order to
provide continuity among cards. Lacing hole was a reference
for the distance calculation. Picture taken from scanner.
Distance calculation based on X and Y co-ordinate value.
Reading from top to middle and from bottom to middle, made
zero bit error.
REFERENCES
[13] Vastrad, J. V, “Weaving computerized negi motifs in traditional
Lakkundi saris”. Ph. D. Thesis, Univ. Agric. Sci., Dharwad (India) 2003
[14] Identification, counting, and sizing of dispersed phase droplet of
scanning electron microscopy micrograph using digital image processing
Phankokkruad,
M.
;
Wacharawichanant,S.
Image and Signal Processing (CISP), 2012 5th International Congress on
Digital Object Identifier: 10.1109/CISP.2012.6469916 Publication Year: 2012
, Page(s): 510 - 514 IEEE Conference Publications
[15] Daggu Venkateshwara “Implementation of evaluation of Image
Processing Algorithm on Reconfigurable Architectural using C-based HDL”
International Journal of Theorotical and applied computer sciences, Las
Vegas ,2006
[16] ENGN 4538 Computer Vision , Semester 1, 2003 Lab 1: Introduction to
Image processing in Mat Lab & Binary Image Analysis.
[1] KanagaRaj “Price Hike of Cotton Yarn and Its Impact on Power Loom
Sector”,National
conference on Emerging challenges for sustainable
Business, IIT Roorkee, 2012
[2] “Introductory Notes Concerning Jacquard Technology” written by
Garth Fletcher for submission to the Complex Weavers Jacquard Study Group
© 2002 by Garth Fletcher JacqCAD International 288 Marcel Road Mason,
NH 03048 USA
[3]
“DIGITIZING CONVENTIONAL PATTERNS OF GUJARAT
EMBROIDERY AND PRODUCT DEVELOPMENT” Thesis submitted to
the University of Agricultural Sciences, Dharwad in TEXTILES AND
APPAREL DESIGNING By SHAMEEMBANU A. BYADGI 2009.
[4] Bains, S. and Bhatti, M. S., 2001, Philosophy of Phulkari. Tex. Trends,
44 (5):
[5] Bains, S. and Bhatti, M. S., 2001, Software for Phulkari design. Tex.
Trends, 44 (7): 25 –26.
[6] Han Antai,Guo Xiahoua “Design of multi system of-function jacquard
knitting machine” Journal of textile research ,chaina, Vol 30, pp111-116, Mar
2009.
[7] Automation of hand embroidery motifs into self woven designs
Karnataka J. Agric. Sci., 23 (4) : (668-672) 2010
[8] R. C. Gonzalez, R. E. Woods, S. L. Eddins, “Digital Image Processing
using MATLAB”, ISBN 81-297-0515-X, 2005, pp. 76-104,142-166
[9] DEDIP: a user-friendly environment for digital image processing
algorithm development.Keddy,W.A.;Agathoklis,P. communications,
Computers and Signal Processing, 1991., IEEE Pacific Rim
Conference on Digital Object Identifier: 10.1109/PACRIM.1991.160844
Publication Year: 1991 , Page(s): 733 - 736 vol.2 Cited by: Patents (24)
IEEE Conference Publications.
[10] Signal & Image Processing : An International Journal(SIPIJ)Vol.1,No.2
, December 2010
[11] Shameembabu A Byadgi “ Digitising Conventional Patterns Of Gujarat
Embroidery and Product development” Thesis submitted to the University of
Agricultural Sciences, Dharwad, 2009.
[12] Karnataka J. Agric “Automation of hand embroidery motifs into self
woven designs”. Sci., 23 (4) : (668-672) 2010
www.ijmca.org
Page 74