Image analysis – Applications – Textiles or clothing
Reexamination Certificate
2000-09-06
2004-05-25
Patel, Jayanti K. (Department: 2625)
Image analysis
Applications
Textiles or clothing
C356S429000
Reexamination Certificate
active
06741726
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to a system and method for rapidly and electronically scanning long lengths of one or more yarns, and displaying measured qualities of a large number of selected lengths of the yarn in a fabric pattern to assist in the evaluation of the effects of the quality variations on the quality of fabric which potentially would be produced from the one or more yarns.
BACKGROUND OF THE INVENTION
There are numerous quality tests for yarns and the fabric they will produce. One such test is for the general appearance of the fabric. This test is accomplished by weaving or knitting the subject yarns into a fabric sample. Yarn quality defects and effects, such as color inconsistencies, width inconsistencies, hairiness, slubs, broken ends, and other novelty effects can thereby be judged as they will affect fabric quality, unarguably the most critical aspect of the yarn. The occurrence of yarn defects in close proximity on the fabric, of undesirable defect patterns, of a multitude of marginally acceptable yarn defects creating an unacceptable fabric defect, can be visually observed and quantified. This test has been standardized as ASTM D2255/90 and ASTM D2255. It has become the accepted method for predicting the quality of fabric to be produced from entire lots of sampled yarns based on the probability that unacceptable fabric defects will be observed in a sample of this size.
Fabric quality will be affected by other factors as well as yarn quality, such as loom defects. Because looms and knitting machines intended for high production runs of fabric cannot practically be interrupted for the running of small test runs, as reasoned below, the machines used for test sample runs tend to be smaller, older, worn, or obsolete models whose results would be generally unacceptable for producing marketable fabric. Variations in tension on one or more yarn strands during weaving or knitting will cause fabric variations that might be read as resulting from poor yarn quality. Sinusoidally occurring loom inconsistencies will cause recurring fabric faults whose regularity may not be recognized in the small test sample, and may therefore be read as resulting from poor yarn quality. Other loom deficiencies and also ambient temperature and humidity conditions may affect fabric quality independently of the quality of the yarn from which it was produced. Obviously, the grading of yarn in fabric samples is a somewhat qualitative process, leaving the possibility that one grader might judge a sample quite differently than another grader, that the subject sample might not accurately represent the fabric to be produced, or that defects will be missed as a result of the small sample size.
Other drawbacks to the current grading system are that it requires the actual weaving or knitting of a fabric sample, or winding the yarn on a board, which requires specialized equipment and specialized operator training, or an actual loom or knitting machine. Test samples can be produced using an instrument such as a Fiber Analysis Knitter (FAK) like that manufactured by Lawson-Hemphill, Inc., of Central Falls, R.I. Looms and knitting machines are huge and complicated machines requiring many manual settings and adjustments independent to each run. These machines cannot practically fit within small quality labs and require operator expertise not typical of lab personnel. A significant amount of the time taken to produce a fabric run is in the set-up of the machine. Each fabric pattern requires a different loom set-up procedure. Looms are quite expensive to purchase and operate, take up substantial floor space, and must be used efficiently in the production of marketable fabric in order to justify their high cost. For reasons of time and cost, the practice of setting-up a production loom to run a few small test samples of a particular fabric would be impractical. Test samples are generally run on smaller weaving or knitting machines which are more practical for this application even though those tend to be slow and to have outputs which may not be fully indicative of production fabric.
Further drawbacks to the current test sample making method is that many actual yarn sample spools are required to run even these small test samples, and the yarn waste percentage is very high on such a short run.
It can readily be seen that the current fabric grading method is very slow, inaccurate, labor intensive, highly judgmental and expensive.
SUMMARY OF THE INVENTION
It is therefore the object of this invention to provide a system and method for electronically displaying simulated fabric including one or more measured qualities of a large number of yarn lengths, measured from one or more yarn samples, to assist in predicting fabric quality as it would be affected by those measured yarn qualities.
The term “display” as used herein is meant to encompass human-viewable computer outputs, such as a CRT or a printer output, and/or a machine-interpretable data set-such as the data from which a visible display could be derived, depending on the possibilities in each given use. As explained herein, the data set is created from measurements, or representations of such measurements, taken along one or more lengths of yarn.
This invention results from the realization that a fabric sample evaluation test may be accomplished electronically by scanning long lengths of one or more yarn samples from one or more yarn cones and arranging and displaying the long lengths of yarn as a number of interwoven shorter lengths in a typical fabric pattern to electronically simulate the manual fabric test sample currently in use, and by representing different measured yarn qualities with display attributes such as their actual image, display line width or length, shades of grey, colors, or numerical or other values, to allow the visual evaluation of the simulated fabric as done manually with fabric samples, and also allow electronic yarn gradings and actual counts of faults or events in the yarn and fabric.
It is a further object of this invention to provide such a system and method that is less expensive to accomplish than the manual system. The need for valuable loom time is virtually eliminated. Fabric simulations will most often be sufficient to judge yarn lot acceptability.
It is a further object of this invention to provide a system for predicting the quality of finished goods made of fabric woven or knitted from a particular yarn sample.
It is a further object of this invention to provide such a system and method that is quicker than the manual system. Yarn measurements can be converted to simulated fabric almost instantly. Loom set-up time and run time is eliminated.
It is a further object of this invention to provide such a system and method that is easier and less labor intensive than the manual system. Once a yarn length is measured, manual labor to provide the simulated fabric display can be reduced to a few keystrokes.
It is a further object of this invention to provide such a system and method that provides results which can be more accurately quantified, and thereby eliminate human judgement. The number of yarn faults on a fabric sample can be readily established. The proximity of yarn faults/qualities, and their relationship within a predetermined fabric zone can be evaluated and quantified or stored. These results can then be compared to an established tolerance or reference to eliminate the subjectivity of the current grading process. One or more simulated fabric images could be displayed simultaneously on a CRT screen to compare one simulated fabric to another, or to a standard. One yarn could be displayed simultaneously woven into one or more patterns, or woven with other yarns.
It is a further object of this invention to provide such a system and method that allows for the grading of a greater quantity of yarn cones and fabric test samples than is currently accomplished. Any possible arrangement of any number of woven or knitted yarns can be accomplished rapidly. Worst-case scenarios can be located and the statis
Bonneau David
Gordon, Jr. Kendall W.
Nevel Avishai
Azarian Seyed
Haeckl, Esq. Jenifer E.
Lawson-Hemphill Inc.
Mirick O'Connell DeMallie & Lougee LLP
Patel Jayanti K.
LandOfFree
System and method for electronically evaluating predicted... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for electronically evaluating predicted..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for electronically evaluating predicted... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3270701