Deep learning algorithm-based quantitative modeling method of near-infrared spectroscopy of tobacco and model application

Abstract

The invention discloses a deep learning-based quantitative modeling method of near-infrared spectroscopy of tobacco. A near-infrared spectrometer is utilized to collect spectroscopy information, the near-infrared spectroscopy information of the tobacco is acquired, and spectroscopy data are preprocessed; main chemical component information of the tobacco is acquired; a sparse feature learning method is used to create an over-complete dictionary by applying the near-infrared spectroscopy data of the tobacco and a K-SVD algorithm, and an OMP algorithm is utilized to calculate and obtain sparse representation coefficients of spectroscopy; and a PSO-SVM learning algorithm is adopted, and the sparse representation coefficients and the chemical component information are combined to establish a near-infrared spectroscopy regression prediction model. According to the method, dual technologies of spectroscopy analysis and machine learning are utilized, and a support vector machine algorithm in pattern recognition is combined to realize fast quantitative modeling for the near-infrared spectroscopy of the tobacco, and the established model is applied to accurately predicting the main chemical component information of the tobacco.

Description

technical field [0001] The invention belongs to the technical field of using near-infrared spectrum to analyze chemical components of tobacco leaves, and in particular relates to a near-infrared spectrum quantitative modeling method for analyzing and predicting chemical components of tobacco leaves and its application. Background technique [0002] Near-infrared spectroscopy has the advantages of simplicity, rapidity, simple pretreatment, non-destructive and non-polluting to samples, and simultaneous determination of multiple components. It is widely used in agriculture, petroleum, tobacco and other fields. The near-infrared spectroscopy method mainly uses the vibration of chemical bonds such as C-H, N-H, O-H, and C-C in organic matter, and the chemical components such as total sugar, total nitrogen, reducing sugar, nicotine, and chlorophyll in tobacco leaves are rich in hydrogen-containing groups. Therefore, the key features contained in the spectrum of tobacco leaves can b...

AI Technical Summary

Problems solved by technology

[0003] The near-infrared spectrum of tobacco leaves has shortcomings such as weak signal strength, spectral bandwidth, overlapping spectral peaks, interference, and the inability to directly extract useful information from the spectrum. Qualitative and quantitative analysis, so the near-infrared spectral modeling of tobacco leaves is the core of tobacco leaf near-infrared spectral analysis technology

Although the existing near-infrared spectral modeling methods can meet the basic application requirements in most cases, the existing methods sometimes have the disadvantage of low accuracy in practical applications, and there is still no method that can be used in all In order to obtain the best results in the application of tobacco leaves, the appropriate method must be selected according to the specific object. Therefore, in order to further improve the accuracy of near-infrared spectrum analysis of tobacco leaves and improve the modeling method of tobacco leaf near-infrared Improve and optimize the modeling method

Images