（CTEI网讯 ）

会议：第十二届亚洲纺织会议暨2013中国纺织学术年会

时间：2013年10月24-26日

地点：上海东华大学

分会场：纺织品性能、测试与评价

报告题目：一种纱线缺陷特性描述和分类的信号处理方法

报告人：V.K. Yadav

报告英文简介：

Spun yarn faults and imperfections arising out of mass variation, exhibit distinct characteristics in terms of their contours, sizes and shapes apart from their frequency of occurrences along the yarn length. The present work is aimed to classify and characterise the imperfections based on their configuration and volume, using a novel mathematical framework. A 30 tex cotton ring spun yarn is used for creating a library of imperfections divided over three groups apart from group for normal yarn. These were digitally captured and converted to one-dimensional signals using Finite Radon Transform after pre-processing of images. By regarding these signals as realisation of a stochastic process, Karhunen–Loève Transformation (KLT) basis is computed for all the classes and projection angles. Eigen values, with the top six highest energy, are used for forming eigen subspaces for each class. Any realisation, of test class, is classified by projecting it on the common KLT basis and using minimum distance criteria. Classified imperfections, under each class, were analysed for their volumetric distribution and sizes, by dividing the imperfection longitudinally in equal bands. Profile of band volume along the imperfection length are obtained along with the total volume of imperfection. A distinct distribution of imperfection is found in terms of their volume and physical dimensions. This opens up a basis of sub-classification of imperfections based on the absolute volume, volume profile and the length.

【专题】第十二届亚洲纺织会议暨2013中国纺织学术年会

http://www.ctei.gov.cn/special/2013nzt/atc12/