In the name of ALLAH the beneficent the merciful

In the name of ALLAH the beneficent the merciful

In the name of ALLAH the beneficent the merciful Noman Haleem B.E. in Textile Engineering from National Textile University Specialization in Yarn Manufacturing Topic Determination of Twist per Inch value in Double Yarns using Digital Image Processing Techniques

Determination of twist per inch value in doubled yarns using digital image processing techniques By, Noman Haleem TPI value and its significance Affects on yarn properties Yarn strength Yarn comfort Moisture absorbency Affects on fabrics Dye uptake Affects on garments serviceability

Abrasion Pilling Aesthetics affects Conventional Methods of TPI determination Manual Testers Automatic testers Direct counting methods Twist untwist method Take up twist tester Conventional TPI tester Limitations of the conventional methods Destructive nature of testing

Offline testing modes Slow speed of testing Manual calculation Human error Machine performance and calibrations Image processing The analysis of a picture using techniques that can identify shades, colors and relationships that cannot be perceived by the human eye. Image processing is used to solve identification problems, such as in forensic medicine or in creating weather maps from satellite pictures. It deals with images in bitmapped graphics

format that have been scanned in or captured with digital cameras Image processing and textiles Image processing is already being used in various branches of textiles for technical measurements and properties evaluation. Contamination sorters Motion sensing purposes Fiber properties evaluation Fabric properties evaluation Color matching and shade variation detection Research sequence

Hardware setup Image analysis Pattern recognition Algorithm design Algorithm evaluation GUI design Proposed hardware design Image analysis 1 Yarn image of 1 inch length taken by a sensitive camera in presence of a predefined

illumination setup. The background was set to black to evident more details of yarn structure during imaging. Background adjustment Intensity adjustment of image IMADJUST The intensity of digital image of yarn is adjusted in order to evident the minute

details more clearly. Image cropping and pattern recognition Phenomenon of alternating color intensities visualized in gray scale image of yarn present in yarn texture due to irregular reflection of light. Based on these bright and dark clusters in the yarn body there is a certain pattern can be observed that repeats itself after a specified number of pixels. Graph between pixel intensity and number of pixels Graph: Relation between intensity values of pixels and no. of pixels

ROI filtering on basis of pixel intensity ROIFILT Figure: Region of interest filtering on basis of intensity values NFILTER Figure: Implementation of neighborhood processing to precise the results Binary pixels graph Figure: Identifiable dark and bright places in binary image of cropped area BWLABEL

Results for 8 double yarns Mean Calculated Twist = 10.4 Actual Twist = 10.76 Sample No. Calculated TPI 1 10 2

10 3 11 4 11.5 5 10 6

10.5 7 10 8 10.5 9 10.5

10 10 Standard Deviation = 0.516 CV %age = 4.96% Results for 10 doubled yarns Mean Calculated Twist = 8.8 Actual TPI = 8.6 Sample No. 1

Calculated TPI 8.5 2 9.5 3 8 4 9

5 8.5 6 8.5 7 9 8 8.5

9 9.5 10 9 Standard Deviation = 0.483 CV %age = 5.49% Results for 15 double yarns Mean Calculated Twist = 13.5

Actual Twist = 13.75 Sample No. Calculated TPI 1 14 2 13 3

13.5 4 13.5 5 12.5 6 14

7 13.5 8 13.5 9 13.5 10 14

Standard Deviation = 0.471 CV %age = 3.49% Conclusion As it is clear from the above calculated values of TPI, comparison of reference values and calculated values, standard deviation from mean value and CV %age of TPI calculated, the image processing technique can be used in the determination of twist per inch values of doubled yarns on commercial level. Although there are variations from the average value and also there are limitations in this image processing algorithm but these variations can be fixed with more precise hardware setup and fine tuned IP algorithm. That means a combination of simple setup of hardware and user friendly software can eliminate the conventional testing

techniques and their drawbacks for measurement of TPI values of yarn. This investigation has opened a new horizon for researchers to make this technique usable on the professional grounds to test TPI values of yarn without application of mechanical factors and to enter the era of non destructive and online testing. References Adams, R. (September, 1993). Radial Decomposition of Discs and Spheres Computer Vision, Graphics, and Image Processing: Graphical Models and Image Processing, Vol. 55, No. 5, pp. 325-332. Cybulska, M. (1999). Assessing Yarn Structure with Image Analysis Methods 1, Textile Res J. 69(5), 369-373. Savialle, B. P. (1999). Physical testing of textiles. (pp. 85-94). England: Woodhead Publishing

Ltd. Schwarz, E.R. (1938). Twist Measurements in Single Yarns, Textile Res. J. 8, 157-164. Young, H. D., & Freedman, R.A. (2000). University Physics with Modern Physics (10th ed.). USA: Addison-Wesley Publishing Company. Matlab image processing toolbox online Documentation. Retrieved May 16, 2008, from Thank you very much for your kind attention

Recently Viewed Presentations

  • Westgard Multirule System

    Westgard Multirule System

    rules to assess if a system is in control use when at least 2 levels of control are run with the examination run cannot use with only one control Dr. Westgard Quantitative QC-Module 7 Westgard Multirule System Titles 12S rule...
  • Postgraduate Award in Professional Skills

    Postgraduate Award in Professional Skills

    Conclusions - include thoughtful logical conclusions that demonstrate reflections on what you have learnt from the audit 8. References - all material should be properly referenced using Vancouver system (BMJ) (preferred) or Harvard formats Reflection on experience (NHS Foundation Learning...
  • A Closer Look at Logos - University of Arizona

    A Closer Look at Logos - University of Arizona

    ENGL102 Ordover Fall 2008 A Closer Look at Logos Syllogism, Enthymeme, and Logical Fallacies What is a syllogism? a specific method of logical deduction (moving from the general to the particular) every syllogism contains at least three parts: a major...
  • William B. Gibbs Jr. Elementary School End of

    William B. Gibbs Jr. Elementary School End of

    Collaborate 10 hours weekly whole group and small group instruction, 2.5 hours weekly instruction provided separately (general education/special education) for literacy block - response to data collection. Data Collection. IEP - data on goals and objectives. Behavior. Organization - user...
  • Writing: Referencing Sources Debra Myhill References What is

    Writing: Referencing Sources Debra Myhill References What is

    Referencing Conventions. There are lots of different styles used to reference other people's work, but the core principles are the same. This presentation focus on the Harvard style, sometimes called the author-date method, as it is the easiest and clearest.
  • Defined as a violently rotating column of air

    Defined as a violently rotating column of air

    Occur in many parts of the world ... Outdoors: Lie down in a depression, ditch or culvert. Cover your head with your arms. (3) In a vehicle or mobile home: In a vehicle or mobile home is probably ...
  • Overview and Reflections on China's Global Role in Dam Building

    Overview and Reflections on China's Global Role in Dam Building

    South-East Asia, South Asia (primarily Pakistan and Nepal), Africa, Latin America and Eastern Europe. Some trends: Regional concentration. What types of dams, what size. New markets ... Aspire to be the biggest and best dam-builders in the world ... Overview...
  • ER Diagram Notation - Indiana University Bloomington

    ER Diagram Notation - Indiana University Bloomington

    Strong vs. Weak entities. Strong Entity = existence-independent entity. Weak Entity. existence-dependent entity in a strong relationship. inherits all or part of its primary key from parent entity. entity w/ clipped corners in CF model, double-walled in Chen model