Study on characteristics analysis and recognition for infrared absorption spectrum of foreign fibers in cotton

A method of using infrared absorption spectrum to detect foreign fibers in cotton is put forward. Based on the analysis of infrared absorption spectrums of cotton fiber and foreign fibers, the characters to distinguish the difference between cotton and foreign fibers are fixed on, and the definition and calculation method of eigenvalues of the fiber are given. Aimed at the recognition of the eigenvalues, a reformative counter-propagation neural network is designed to identify foreign fibers from cotton. The experiments indicated that the way and method described in the paper are accurate and efficient.