Wood Stain Formulation Prediction with a CNN-GRU Network Based on LDA and Enhanced Cuckoo Search Algorithm
To accurately predict wood dye color matching formulations,a hybrid neural network model combining linear discriminant analysis(LDA),improved cuckoo search(ICS)algorithm,convolutional neural network(CNN),and gated recurrent unit(GRU)was proposed.The model processed and classified the spectral information by LDA for dimensionality reduction,extracted essential features utilizing CNN,input these characteristics into GRU for training,and optimizeed the hyperparameters in the network using the ICS algorithm.The model's performance was measured through various evaluation criteria,including the coefficient of determination R2 and the color difference calculation formula(CIEDE2000).In comparison with multiple traditional models,the proposed model demonstrates superior performance.Additionally,the model has a relatively low number of parameters,high computational efficiency,and excellent stability and reliability.The results show that the proposed model exhibits significant advantages when applied to predicting wood dye color matching formulations based on spectral information.
deep learningwood dyeingimproved cuckoo search(ICS)
万方数据
维普
