Spectral classification method for excellent high-quality milk, high-protein special milk, high-cream special milk and ordinary milk

A grading method and high-protein technology, which is applied in the direction of material analysis, material analysis, character and pattern recognition through optical means, can solve the problems of redundant spectral information and unclear spectral characteristic wavelengths, and improve model accuracy, The effect of fast identification speed and saving instrument cost

Pending Publication Date: 2021-08-31
HUAZHONG AGRI UNIV
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, for milk with different milk fat percentages, milk protein percentages and somatic cell counts, no research has been done on rapid quality grading.
[0004] In addition, these studies still have many redundant information in the spectrum, and the spectral characteristic wavelength is not clear.

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
  • Spectral classification method for excellent high-quality milk, high-protein special milk, high-cream special milk and ordinary milk
  • Spectral classification method for excellent high-quality milk, high-protein special milk, high-cream special milk and ordinary milk
  • Spectral classification method for excellent high-quality milk, high-protein special milk, high-cream special milk and ordinary milk

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] Example 1: Establishment of spectral rapid grading method for premium quality milk, high protein specialty milk, high milk fat specialty milk and ordinary milk

[0042] (1) Test materials and methods

[0043] A total of 5121 milk samples from 10 different pastures in Hebei Province, China were selected and numbered. Pour the sample into a cylindrical sample tube with a diameter of 3.5cm and a height of 9cm, bathe in a water bath at 42°C for 15-20min, and use MilkoScan from FOSS Company TM 7RM milk composition detector, extend the solid optical fiber probe into the liquid, scan the sample, and obtain the milk protein, milk fat content and mid-infrared spectrum of the milk sample (see figure 1 ), using Fossomatic from FOSS TM 7 The somatic cell detector was used to measure the number of somatic cells in milk, and the composition differences of the four kinds of milk are shown in Table 1.

[0044] Table 1 Differences in milk components

[0045]

[0046] (2) Selecti...

Embodiment 2

[0059] Example 2: Effect of wavenumber on model accuracy

[0060] The mid-infrared spectrum collection of milk, method is the same as embodiment 1, removes 1597-1712cm -1 and 3024-3680cm -1 After the wave number, analyze 3680-4000cm -1 Effect of wavenumber on the model. Table 3 shows that the removal of 3680-4000cm -1 After wave number, the training set accuracy and test set accuracy of the model are both improved, which are 84.4996% and 84.2243% respectively. So finally choose 925-1597cm -1 and 1712-3024cm -1 The combination of bands is used to build the model as a full spectrum.

[0061] Table 3 Effect of wave number on model accuracy

[0062]

Embodiment 3

[0063] Example 3: Effect of derivative processing on model accuracy

[0064] The method for collecting the mid-infrared spectrum of milk is the same as in Example 1, and the mid-infrared spectrum value of milk is preprocessed by using the first order derivative and the second order derivative respectively. The number of smoothing points is set to 7, 9, and 11 respectively, and the preprocessed spectrum is brought into the NB model. The results are shown in Table 4. Table 4 shows that the NB model established after the 9-point smooth derivative preprocessing is the best, the accuracy rate of the training set and the accuracy rate of the test set of the first derivative are 90.6886% and 88.8527%, respectively, and the accuracy rate of the training set of the second derivative is 88.8527%. The test set accuracies are 93.7552% and 92.0469%, respectively. Therefore, 9 is selected as the best number of smoothing points, and the first and second derivatives of 9-point smoothing are ...

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 belongs to the technical field of dairy product analysis and particularly relates to a spectrum grading method for excellent high-quality milk, high-protein special milk, high-cream special milk and common milk. The method is related to the field of mid-infrared spectrum analysis. The method is characterized in that the optimal modeling wave band is the wave band of 925-1597 cm<-1> and the optimal modeling wave band of 1712-3024 cm<-1>; the method comprises the following steps of (1) taking a milk sample to be detected as a detection sample; (2) collecting mid-infrared spectrum data; (3) selecting a spectral band; (4) preprocessing the original mid-infrared spectrum data; (5) extracting a mid-infrared spectrum characteristic wavelength; (6) predicting samples in the test set by using the model; (7) comparing and evaluating the model; and (8) through comparative analysis, selecting an optimal grading model. The method is advantaged in that the model is established by utilizing the characteristic variable combination, so the model is simplified, and model precision and the detection speed are improved.

Description

technical field [0001] The invention belongs to the technical field of milk product detection, and in particular relates to a spectral identification method for super-high-quality milk, high-protein characteristic milk, high-butter fat characteristic milk and common milk. The field of the invention is related to the field of mid-infrared spectroscopy. Background technique [0002] Special milks such as "high protein" and "high milk fat" generally appear on the market. In addition, the number of somatic cells in milk has a significant impact on milk protein and milk fat content [1] . Therefore, somatic cell count is also an important indicator for evaluating milk quality. If the quality of milk can be accurately graded in batches, the production efficiency and economic benefits of dairy enterprises will be greatly improved. [0003] Traditional chemical analysis methods for milk protein, milk fat and somatic cell count are time-consuming and polluting the environment. If...

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): G01N21/3577G06K9/00G06K9/62
CPCG01N21/3577G01N2021/3595G06F2218/04G06F2218/08G06F2218/12G06F18/24155
Inventor 张淑君肖仕杰王巧华李春芳王海童马亚宾倪俊卿张依罗雪路樊懿楷
Owner HUAZHONG AGRI UNIV
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