Research On Hyperspectral Wood Dyeing And Color Matching Algorithm Based On Particle Swarm Optimization

Wood color is an important factor in determining the quality of wood,and because of the scarcity of precious wood,so derived simulation wood processing industry,the traditional wood dyeing industry production line workers rely only on experience and visual judgment for dyeing,resulting in the decline of dyeing quality,waste of wood resources,and failure to achieve the desired dyeing effect.To address the above problems,this study applies hyperspectral imaging technology and computer intelligent algorithm to wood dyeing color matching to improve the accuracy and practicality of wood dyeing computer intelligent color matching application.The main research contents of this paper are as follows.(1)The research takes ash veneer wood as the research object to conduct wood dyeing test under the determined dyeing process,and the experiments obtain precious species wood samples,single component dye dyed wood samples and mixed dye dyed wood samples,and elaborate the principle of hyperspectral imaging technology and several color space and color difference rating theories.For the experimental study of wood dyeing color matching theory,hyperspectral instruments were selected to handle the experimental samples,and hyperspectral techniques were used to collect,process and analyze the data of the experimental samples.(2)Based on the training sample set of experimental data,a prediction model of wood dyeing formulation based on Kubelka-Munk theory was established.The optimized Friele algorithm model was established for the experimental study of wood dyeing and color matching to address the problems of the Kubelka-Munk theory algorithm model,and the color difference accuracy of the Friele algorithm model before and after the optimization was compared in the process of wood dyeing and color matching prediction formulation.Using hyperspectral instruments to detect the dyeing experimental samples,the particle swarm optimized Stearns-Noechel algorithm model was used to predict the recipe based on the comparison of the dyeing effect and the algorithm,in order to improve the dyeing accuracy of the Friele algorithm model,reduce the color difference and spectral reflectance curve error,and compare the Stearns-Noechel algorithm model before and after the optimization in the process of wood dyeing and color matching The accuracy of color difference in the process of predicting the recipe.This paper adopts the research of wood dyeing color matching based on hyperspectral imaging,which transforms from the traditional color matching method to computerized intelligent color matching,realizes the real-time monitoring of wood dyeing process,corrects the dyeing deviation in time,improves the quality of wood dyeing,and liberates the surplus labor,saves the production resources,reduces the production cost,and at the same time,greatly improves the product qualification rate as well as the production efficiency,and improves the accuracy and precision of wood dyeing color matching.