Research On The Computer Intelligent Color Matching Technique For Wood Dyeing Based On Hybrid Model

Wood dyeing is used to improve the value of wood products and the efficiency of wood utilization.This research is significant to improve the efficiency of wood dyeing and accelerate the industrialization process.The establishment of intelligent color matching mixed model for wood dyeing enriches the theoretical basis of computer intelligent color matching.Research on intelligent hybrid model of wood dyeing color matching based on dye formulation can improve production efficiency and reduce production costs,which is conducive to the formation of the wood color dyeing production specifications.According to the optics and colorimetry characteristic of color,the correspondence between the chromatic aberration before and after dyeing and the concentration values of the dyestuffs is confirmed through a lot of experiments.The peak density function,wavelet transform,genetic algorithm are integrated into RBF neural network to construct an optimal mixture model of wood dyeing.Conclusions are achieved as follow:(1)The peak density function is used to determine the number of neurons in the hidden layer of the RBF neural network.The centers and the widths of the radial basis function of the hidden layer nodes were initialized by extracting the features of samples.The results proved that the average relative error is only 0.62% in 50 epochs.The convergence rate and precision were improved significantly by the using of the peak density function.(2)The wavelet function is used to be the basis function of hidden layer neurons,which expressing strong localization characteristics in time domain and frequency domain.The results proved that the average relative error is only 0.59% in 38 epochs.The local optimization ability of RBF neural network is enhanced while weakening the generalization ability of RBF neural network.(3)Genetic algorithm is proposed to optimize the centers and the widths of hidden nodes and the connection weights between hidden layer and output layer of RBF neural network globally.The average relative error of the RBF neural network which is optimized globally by genetic algorithm is only 0.42% in 20 generations.Each generation group contains the whole parameters of RBF neural network.(4)MCGS configuration software is used to simulate the computer automatic color matching system.Computer control system is used to collecting and monitoring the on-site data,which can provide complete statistics with perfect function,simple operation,good visibility and strong maintenance.