Research on Spectral Region Selection of Near Infrared Spectra Based on Genetic Algorithm

In recent years, the spectral regions optimization of near infrared spectroscopy using the intelligent computing algorithms has achieved good application results. According to the spectrum optimization problem of near infrared spectroscopy, the near infrared spectral regions of tobacco samples are optimized using genetic algorithm in this stduy. Simultaneously, the genetic algorithm is combined with partial least squares method to establish the quantitative analysis models of 250 tobacco samples, using 80 tobacco samples to detect the predictive ability and generalization ability of the model. The experiment results show that the genetic algorithm, combined with the partial least squares method is satisfactory. Moreover, the two indicators, the correlation coefficient (R
²
) and root mean square error of prediction (RMSEP) of the quantitative analysis models show satisfactory results. The research provides theory reference and technical support for the infrared spectral region optimization and quantitative analysis.