The research on identification of wool or cashmere fibre based on the digital image

This paper mainly studies computer automatic measurement of cashmere diameter, and the method of cashmere fiber. Simulation experiment is carried out under the simulation environment of Matlab7.0 and the processing scheme which is suitable for the fiber images is confirmed, and finally, on the basis of the images after pre-processed, automatic measurement of cashmere diameter is achieved by the arithmetic of subsection measurement. Experimental results show that measurement methods used in this paper improves the accuracy of measurement of cashmere diameter to some extent. The model features of fiber scale is the based primarily of identification of animal fiber, this paper studies the four kinds of cashmere and wool, six characteristics of fiber diameter, the diameter of high ratio, scale height, scale projection width, scale thickness at right angles and scales diameter difference, and then get the distribution of these six characteristics, identify it by using Bayesian model finally. The results show that the method is very effective to identifying between common wool and cashmere and between wool with stretch wool general.