Tobacco leaf mildew rapid identification method based on wavelet transform-random forest algorithm

A random forest algorithm and wavelet transform technology, applied in character and pattern recognition, calculation, computer parts and other directions, can solve the problem of not considering tobacco leaf pre-judgment, and achieve effective recognition and accurate prediction, high recognition rate and forecast rate. Effect

Inactive Publication Date: 2019-10-22
CHINA TOBACCO GUANGDONG IND
View PDF8 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Based on the change of ergosterol content in tobacco leaves, Zhou Jiyue[10] established a pattern recognition model for the mildew process of tobacco leaves by using near-infrared spectroscopy combined with chemometric methods. Changes in the content of ergosterol in the medium can better exclude the influence of differences in tobacco leaf regions, parts and grades on the model, but because of this, there are bound to be limitations in the transfer and application of the model. , need to rebuild the model
Yang Lei [11] collected near-infrared spectra of tobacco leaves in the range of 780nm to 2500nm with a near-infrared spectrometer, and established a least squares-discriminant analysis model, which was applied to the prediction of moldy tobacco leaves, but the model only Judgment is made between mildewed tobacco samples and normal tobacco samples, without considering the prediction of tobacco leaves that are close to mildew

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Tobacco leaf mildew rapid identification method based on wavelet transform-random forest algorithm
  • Tobacco leaf mildew rapid identification method based on wavelet transform-random forest algorithm
  • Tobacco leaf mildew rapid identification method based on wavelet transform-random forest algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] Such as figure 1 As shown, the rapid identification method of tobacco leaf mildew based on wavelet transform-random forest algorithm includes the following steps:

[0037] S1: collecting tobacco leaf samples;

[0038] S2: collect the near-infrared spectrum of each tobacco leaf sample separately, as the spectral information of each tobacco leaf sample;

[0039] S3: Use discrete wavelet transform to preprocess spectral information, and solve to obtain wavelet coefficients;

[0040] S4: Use the random forest algorithm to identify the mildew degree of the tobacco leaves from the wavelet coefficient, and complete the identification of the mildew of the tobacco leaves.

[0041] More specifically, the step S2 is specifically:

[0042] S21: Put the collected tobacco leaf samples into the sample cups respectively, collect the infrared spectrum data of each tobacco leaf sample respectively, and use it as the basic spectral information of each tobacco leaf sample;

[0043] S22...

Embodiment 2

[0057] More specifically, on the basis of Example 1, the mildewed tobacco samples were prepared. Put the re-cured tobacco leaf samples in an environment with a temperature of 22±2°C and a humidity of 60±5% to balance for 48 hours, put the balanced samples into a constant temperature and humidity box, and adjust the temperature and humidity to 25°C and 85% respectively. For the mildew test, take 40 days as a cycle, and take samples in the following way: on the 0th day, that is, before putting it into a constant temperature and humidity box at 25°C and 85% humidity, take the first sampling; on the 3rd to 9th days, take The 2nd to 4th samples were taken at intervals of 3 days; the 5th to 12th samples were taken at 2-day intervals at 11th to 25th days.

[0058] In the specific implementation process, 116 kinds of single-material tobacco were sampled 10 times at different stages, and 1160 tobacco leaf samples with different degrees of mildew were obtained; the 1160 tobacco leaf sam...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a tobacco leaf mildew rapid identification method based on a wavelet transform-random forest algorithm. The method comprises the following steps: collecting a tobacco leaf sample; respectively collecting the near infrared spectrum of each tobacco leaf sample as the spectral information of each tobacco leaf sample; preprocessing the spectral information by using discrete wavelet transform, and solving and obtaining a wavelet coefficient; and identifying the mildew degree of the tobacco leaves from the wavelet coefficients by utilizing a random forest algorithm to finish the identification of the mildew of the tobacco leaves. According to the tobacco leaf mildew rapid identification method based on the wavelet transform-random forest algorithm provided by the invention, a rapid identification method of tobacco leaf samples with different mildew degrees is established by using a near infrared spectrum method, and a basis is provided for early warning of tobacco leafmildew; spectral data is processed through wavelet transformation, spectral variables are determined, a random forest recognition model is established, the recognition rate and the forecasting rate are high, and effective recognition and accurate prediction of the mildew degree of the tobacco leaf sample are achieved.

Description

technical field [0001] The invention relates to the technical field of tobacco mildew identification, and more specifically, relates to a rapid identification method for tobacco mildew based on wavelet transform-random forest algorithm. Background technique [0002] Tobacco leaf is a special leaf plant, and tobacco storage is an extremely important link. Tobacco leaves grown in the field are harvested, roasted, purchased, transported, threshed and re-roasted, and alcoholized before entering the silk-making workshop of the cigarette factory and rolling them into cigarettes. It takes at least 1 to 2 years from tobacco leaves to cigarettes. Mold is the most widely distributed fungal microorganism in nature. As long as it meets the temperature and humidity conditions suitable for its growth and reproduction, it will grow and reproduce rapidly, causing mildew and deterioration of tobacco leaves. Therefore, research on tobacco mildew prevention has important economic value and a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G01N21/3563G01N21/359
CPCG01N21/3563G01N21/359G06F2218/08G06F2218/12G06F18/24323
Inventor 赖燕华陶红林云王予周瑢欧阳路斯林宝敏
Owner CHINA TOBACCO GUANGDONG IND
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap