Deakin University
Browse

File(s) under permanent embargo

Identification and classification of animal fibres using artificial neural networks

journal contribution
posted on 2001-01-01, 00:00 authored by Fenghua SheFenghua She, S Chow, B Wang, Lingxue KongLingxue Kong
It has been an important and challenging task to classify and evaluate the contents in wool blends. Quantitative characterisation of animal fibre scale patterns has attracted considerable attention, since it is the major evidence for identification and subsequent classification purpose. Although techniques such as imaging processing and linear demarcation functions have been used to identify unknown fibre type with some success, a more comprehensive approach is required to perform this task. In this paper, a new approach is presented, which employs non-linear demarcation functions by using an artificial neural network (ANN). Based on scale pattern features extracted by using image processing techniques the artificial neural network (ANN) model is to classify mohair and merino fibres. It is observed that the techniques developed in this work are very effective and have the potential to be applied to other animal fibres.

History

Journal

Journal of textile engineering

Volume

47

Issue

2

Pagination

35 - 38

Publisher

Nihon Sen'i Kikai Gakkai

Location

Osaka-shi, Japan

ISSN

1346-8235

Language

eng

Publication classification

C1.1 Refereed article in a scholarly journal

Copyright notice

2001, Nihon Sen'i Kikai Gakkai

Usage metrics

    Research Publications

    Categories

    No categories selected

    Keywords

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC