Research On The Intelligent Color Matching System For Wood Dyeing Based On The Improved Neural Network

When wood is used in real life,there is often the problem that the surface properties of the wood do not meet human visual requirements,whiclh inevitably affects the use value of wood.This problem is usually solved by a wood dyeing process.In the social production,the wood dyeing color matching process is inefficient and has poor accuracy,the establishment of the dyeing color matching model is difficult,and there is uncertainty in the process of wood dyeing and color matching.In order to solve the above a series of problems,this paper introduces the intelligent algorithm into the wood coloring computer automatic color matching system,which reduces the time required for the color matching process,and the dyeing effect can also meet the expected requirements.The result of this research has played an important role in promoting the development of the wood dyeing process.Moreover,with the introduction of intelligent algorithms,new ideas have been introduced for the establishment of color matching models for wood dyeing,which has played a role in promoting the development of intelligent algorithms.In this study,the RBF neural network was first introduced into computer color matching,and a wood dyeing color matching model was established.The network convergence speed of the model is slow,the dyeing precision is difficult to meet the expected requirement,the adaptability to fresh samples is low,and the possibility of falling into the local optimal value is greater at the same time.In view of these shortcomings,based on the traditional RBF neural network model,a fuzzy algorithm and a genetic algorithm are introduced,and a mixed model of wood dyeing and color matching is established.The built model is implemented through Matlab programming.The simulation results show that the network convergence speed of the model based on the RBF neural network is 158 steps,and the relative error is 1.55%.This model requires a long time for color matching,witch affects the color matching efficiency.Due to the large error,the color matching effect is difficult to meet the target requirements.The network convergence speed of the model based on the fuzzy RBF neural network is 134 steps,and the relative error is 0.96%.Compared with the color matching model based on RBF neural network,the time required for color matching is relatively shorter,and the color matching effect is relatively better.Based on RBF neural network optimized globally by genetic algorithm color matching model,the network is gradually stable when the genetic algorithm evolved to around 20th generation,and the relative error is 0.87%.Tlie time-consuming and effect of color matching are better than the above two models,and this model can meet the target requirements.Comprehensive comparison of three kinds of model,because the color matching model based on RBF neural network optimized globally by genetic algorithm has the fastest convergence speed and the highest precision,this network was selected as the core of the wood dyeing computer intelligent color matching system.Based on this,the computer intelligent color matching system for wood dyeing is designed and completed.