An non-destructive analyzing system was developed automatically, and the applicability of this system was discussed.The discrimination of the stitch shapes is very important for analysis of the stitch construction of knitted fabrics. To discriminatethe stitch shapes, a computer image processing and a pattern recognition were used in the system. The surfaceimage of knitted fabrics was captured with an image scanner. A threshold processing and a histogram processing wereused to determine the locations of knit stitches. After detection of stitch regions, the stitch shapes was discriminated usinga template matching method. For knitted fabrics in which both back and face loops are used such as links-links andtransfer stitches, this system presented in this study can be applied to the analysis of the discrimination of stitch shapes.

View full abstract

In textile, the patterns of the surface may be due to either the yarn's shapes or the repetition of the warp and weftyarn. The regular periodic nature of many textile patterns permits Wavelet transform techniques in image processing tobe used to measure their visual characteristics. The Fourier power spectrum provides a useful description of the spatialfrequency content in a digital image, and in particular the coarseness of any texture present, but loses the local information.The Wavelet transform can obtain both the frequency information and the local information. The Wavelet transformfixed by frequency of fabric density can graphically describe the irregular part of frequency components on textile surfaceand gives a means of extracting the textile defects.

View full abstract

The purpose of this research is to propose the new system, which focuses on detecting abnormal vibration during high-speed winding . When abnormality is detected in the vibration of a winder, it. gives a warning when the abnormality hasreached to the certain level, or stops only the winder with the problem before the winding yarn package is broken tomake a sudden, abnormal stop.
The research analyzes the data obtained from the system as well as the overhauled winders, and makes comparisonwith the data acquired from the winders that were re-assembled and re-adjusted. The results of the analysis have validatedthe effectiveness and practicability of the system that systematically detects the level of abnormality from thewinders to prevent a costly accident. And it is possible to adopt the Preventive Maintenance of winders by the ConditionBased Maintenance System based on this system.

View full abstract