Authors: Xian-Jun Shi, Wei-Dong Yu
Addresses: College of Science, Wuhan University of Science and Engineering, Wuhan, 430073, China; Textile Materials and Technology Lab, Donghua University, Shanghai, 200051, China. ' Textile Materials and Technology Lab, Donghua University, Shanghai, 200051, China
Abstract: The scale patterns of cashmere and fine wool are different and that is a major reference distinguishing them from each other. A technique commonly used consists of analysing their SEM image for detecting cuticle scale edge height (CSH) of fibre. However, the method is expensive and has an average error of 8%. In this paper, a lower cost method is presented. The microscopic images of fibre captured by CCD camera are transformed into skeletonised binary images only having one pixel wide. Four shape parameters of fibre scale are measured and a LVQ neural network model including these measuring data is developed to classify the two type fibres. The simulation testing results show that whether on training set or testing set, the model can always distinguish cashmere from fine wool (70s) effectively and the average classification accuracy are higher than 93%.
Keywords: intelligent classification; artificial neural networks; animal fibre classification; identification; scale pattern; LVQ neural networks; cashmere; fine wool; modelling; simulation.
International Journal of Modelling, Identification and Control, 2011 Vol.12 No.1/2, pp.107 - 112
Published online: 31 Dec 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article