Construction method of cigarette classification and identification model based on near infrared spectroscopy and OPLS-DA

Abstract

The embodiment of the invention provides a construction method of a cigarette classification and recognition model based on near infrared spectroscopy and OPLS-DA, and the method comprises the steps:selecting a to-be-detected sample, and maintaining the sample according to a preset maintenance step; performing spectral scanning on the sample to obtain a near infrared spectrum of the sample, dividing the near infrared spectrum into a calibration set and a test set, processing the near infrared spectrum through different types of spectrum preprocessing schemes, and performing OPLSDA analysis onthe calibration set to obtain a cigarette classification and identification model and the identification accuracy of the calibration set; substituting the test set into the cigarette classification and identification model to obtain the identification accuracy of the test set, determining a spectrum preprocessing scheme meeting a preset requirement according to the identification accuracy, and obtaining the cigarette classification and identification model corresponding to the spectrum preprocessing scheme meeting the preset requirement. By the adoption of the method, the high-accuracy construction method for the tobacco classification and recognition model can be provided, and the tobacco types can be recognized more accurately in the follow-up process.

Description

technical field [0001] The invention relates to the technical field of tobacco component analysis, in particular to a method for constructing a cigarette classification recognition model based on near-infrared spectroscopy and OPLS-DA. Background technique [0002] The style and characteristics of cigarettes are an important part of the quality characteristics of cigarette products, the core competitiveness of cigarette brands, and the main symbol to distinguish cigarette brands. Finished cigarettes mainly maintain the quality and style characteristics of cigarette brands by blending tobacco leaf formula. For a long time, the quality and style characteristics of cigarettes have been judged and identified mainly through methods such as chemical composition of shredded tobacco, mainstream smoke and sensory quality evaluation. [0003] In recent years, in the tobacco industry, near-infrared spectroscopy has been used for quantitative analysis such as the determination and moni...

AI Technical Summary

Problems solved by technology

[0003] In recent years, in the tobacco industry, near-infrared spectroscopy has been used for quantitative analysis such as the determination and monitoring of various chemical components in tobacco. However, when using near-infrared spectroscopy to analyze tobacco, due to the complexity of tobacco components, it is not currently possible There is no definite analytical method to judge the components and types of tobacco with high accuracy

Images