CN111638246A - Soy sauce classification method based on self-made electronic nose system - Google Patents

Soy sauce classification method based on self-made electronic nose system Download PDF

Info

Publication number
CN111638246A
CN111638246A CN202010516889.4A CN202010516889A CN111638246A CN 111638246 A CN111638246 A CN 111638246A CN 202010516889 A CN202010516889 A CN 202010516889A CN 111638246 A CN111638246 A CN 111638246A
Authority
CN
China
Prior art keywords
soy sauce
sample
electronic nose
data
method based
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN202010516889.4A
Other languages
Chinese (zh)
Inventor
钮永莉
刘青
武斌
刘东旭
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chuzhou Vocational and Technical College
Original Assignee
Chuzhou Vocational and Technical College
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chuzhou Vocational and Technical College filed Critical Chuzhou Vocational and Technical College
Priority to CN202010516889.4A priority Critical patent/CN111638246A/en
Publication of CN111638246A publication Critical patent/CN111638246A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Chemical & Material Sciences (AREA)
  • Electrochemistry (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Probability & Statistics with Applications (AREA)
  • Health & Medical Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

The invention discloses a soy sauce classification method based on a self-made electronic nose system. The method is applied to the cluster analysis of the data of the electronic nose of the five soy sauce types and is compared with the traditional GK fuzzy algorithm and the FCM algorithm. The clustering result shows that the M _ GK algorithm has fewer iteration times and higher clustering accuracy; the electronic nose technology is combined with PCA, LDA algorithm and M _ GK clustering, so that accurate classification of soy sauce varieties can be realized, and an effective model for realizing soy sauce classification and identification is provided.

Description

Soy sauce classification method based on self-made electronic nose system
Technical Field
The invention relates to the technical field of information, in particular to a soy sauce classification method based on a self-made electronic nose system.
Background
The soy sauce originates from China, has unique fragrance and various nutritional ingredients, such as amino acids, saccharides, organic acids, minerals, esters, vitamins and the like, is very beneficial to human bodies, and is popular in daily life of people. The taste of soy sauce is an important index of soy sauce quality and quality, and is generally judged by depending on the sense of a quality control expert, but the taste of soy sauce is greatly influenced by subjective factors, and a modern instrument is required to replace or assist the work of the quality control expert. The electronic nose technology is a simple, rapid and objective smell detection technology, comprises knowledge of a plurality of subjects such as sensors, mode recognition, signal processing and the like, and is widely used for analyzing various gas components containing organic volatile matters. The electronic nose has the advantages of high speed, high sensitivity and nondestructive detection, thereby being applied to the fields of food classification and identification and food safety and becoming a research hotspot. For example, the method is widely applied to tea classification detection, meat detection, wine detection and fruit and vegetable detection.
Most of the current electronic nose application research adopts finished electronic nose systems, such as PEN3, INose and the like, which are expensive and have limited market popularization. The invention researches and designs an electronic nose system for soy sauce detection, which firstly uses Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) to reduce the dimension of data, then uses an improved GK clustering algorithm to process an electronic nose signal, and carries out a comparative test with an FCM algorithm and a traditional GK algorithm to realize the detection and classification of soy sauce varieties.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a soy sauce classification method based on a self-made electronic nose system.
Electronic nose system connection
The data acquisition card uses NI USB 6002, manufactured by National Instruments (NI) Inc., which provides eight analog input channels, and interacts with the PC using NI's configuration management software NIMax. The devices of each part of the electronic nose are orderly connected, the sensor array is connected with a data acquisition card, the output end of the data acquisition card is connected with a PC through a USB line, and the power supply is switched on.
In order to achieve the purpose, the invention adopts the following technical scheme:
a soy sauce classification method based on a homemade electronic nose system comprises the following steps:
s1, opening the bottle cover of the sample, and electrifying the electronic nose for 10 minutes for preheating;
s2, putting a 15 ml soy sauce sample into a sample bottle, connecting an air pipe of the sample bottle with an air chamber, placing the sample bottle in the center of the air chamber, and standing the sensor for 60 minutes until the gas is completely volatilized;
s3, collecting and recording data by using an upper computer program compiled by NI LabVIEW 2018, collecting the data once every 5 minutes for three times in total, and taking the average value of the three collection results as stable data of a soy sauce sample;
and S4, after the soy sauce sample is collected once, opening the box cover and standing for 10 minutes to enable each sensor to recover the initial state, and then repeatedly collecting other soy sauce samples. The state of the air chamber can be observed through a LabVIEW response curve, and after the air chamber is completely emptied, data of another soy sauce sample is collected;
steps S2-S4 are repeated until all data acquisitions are complete.
The invention has the beneficial effects that:
1. improvement to GK clustering (M _ GK)
The GK clustering is a fuzzy clustering algorithm, and the algorithm can detect clusters with different shapes in a data set and realize effective classification of the data set. However, the algorithm also has certain disadvantages, such as sensitivity to initial clustering centers and clustering numbers, easy falling into local optimum and the like. The invention uses a method based on density function to determine the initial clustering center, namely, the data area of each sample point is fully considered, firstly, the point with the highest density index is selected as the first clustering center, then the density index values of other sample points are correspondingly adjusted according to the distance between the other sample points and the clustering center, so that the closer points in the sample are, the faster the density is reduced, the cycle is carried out in sequence until enough clustering centers are found.
2. Faster speed and higher precision
The method is applied to the cluster analysis of the data of the electronic nose of the five soy sauce types and is compared with the traditional GK fuzzy algorithm and the FCM algorithm. The clustering result shows that the M _ GK algorithm has fewer iteration times and higher clustering accuracy; the electronic nose technology is combined with PCA, LDA algorithm and M _ GK clustering, so that accurate classification of soy sauce varieties can be realized, and an effective model for realizing soy sauce classification and identification is provided.
Drawings
FIG. 1 is a flow chart of the operation of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
Soy sauce classification method based on self-made electronic nose system
1. The electronic nose system for soy sauce mainly comprises a gas chamber (where a sensor array is arranged), a sample bottle, a gas sensor array, a data acquisition card, a computer and the like.
2. Electronic nose system connection
The data acquisition card uses NI USB 6002, manufactured by National Instruments (NI) Inc., which provides eight analog input channels, and interacts with the PC using NI's configuration management software NIMax. Orderly connecting all parts of the electronic nose, connecting the sensor array with a data acquisition card, connecting the output end of the data acquisition card to a PC (personal computer) through a USB (universal serial bus) line, and switching on a power supply; the gas in the sample bottle enters the gas sensor array, the gas sensor reflects the concentration and the components of the gas through the strength of the sensed electric signal, and the gas mode is identified by connecting the data acquisition card with a computer. The working environment temperature of the electronic nose system is 20-25 ℃, and the humidity is 35-70%.
3. Step (ii) of
S1, opening the bottle cover of the sample, and electrifying the electronic nose for 10 minutes for preheating;
s2, putting a 15 ml soy sauce sample into a sample bottle, connecting an air pipe of the sample bottle with an air chamber, placing the sample bottle in the center of the air chamber, and standing the sensor for 60 minutes until the gas is completely volatilized;
s3, collecting and recording data by using an upper computer program compiled by NI LabVIEW 2018, collecting the data once every 5 minutes for three times in total, and taking the average value of the three collection results as stable data of a soy sauce sample;
and S4, after the soy sauce sample is collected once, opening the box cover and standing for 10 minutes to enable each sensor to recover the initial state, and then repeatedly collecting other soy sauce samples. The state of the air chamber can be observed through a LabVIEW response curve, and after the air chamber is completely emptied, data of another soy sauce sample is collected;
steps S2-S4 are repeated until all data acquisitions are complete.
4. Processing and classifying the collected data
4.1 data dimension reduction
After the data normalization processing, performing 1 st dimensionality reduction on the preprocessed data by using a Principal Component Analysis (PCA) method to achieve the purposes of data dimensionality reduction and redundant information removal, and then performing feature extraction on the dimensionality reduced data by using a Linear Discriminant Analysis (LDA) method to realize dimensionality reduction again. After PCA + LDA operation, the data dimension is reduced to 4 dimensions.
4.2 improved GK clustering for classification
4.2.1 initial clustering center implementation step
1) For sample set X ═ X1,x1,…,xN) Each sample point x inkDefining a density function
Figure BDA0002530453570000051
Wherein r isaIf the value is more than 0, the value is the neighborhood radius, the sample points outside the neighborhood radius contribute very little to the density index of the point, and the point with the maximum value is selected
Figure BDA0002530453570000052
As the first cluster center.
2) Assume that the k-th selected cluster center is
Figure BDA0002530453570000053
Corresponding to a density function of
Figure BDA0002530453570000054
The density function for the other sample points is given by the following formula:
Figure BDA0002530453570000055
correcting to select the point with the highest value as the clustering center rb＝1.2raTo 1.5rb。
3) Judgment of conditions
Figure BDA0002530453570000061
And (3) if the judgment is not true, repeating the step (2), and if the judgment is true, exiting. η is a predetermined value, and generally η is 0.5.
4.2.2 improved GK clustering algorithm
The improved GK clustering algorithm comprises the following steps:
determining a clustering number c, a fuzzy index m and a final allowable error;
c initial clustering centers are determined by the step of 4.2.1, and an initial value of a fuzzy partition matrix U is set;
updating the clustering center according to the formula (4);
Figure BDA0002530453570000062
calculating covariance matrix F of ith clustering centeriComprises the following steps:
Figure BDA0002530453570000063
calculating a positive definite symmetric matrix A according to the formula (6)iThen solving for x according to equation (5)jTo the center of the cluster viDistance norm of
Figure BDA0002530453570000064
Figure BDA0002530453570000065
Figure BDA0002530453570000071
Updating a fuzzy partition matrix U by using a formula (7);
Figure BDA0002530453570000072
for a given > 0, when | | | U is satisfied(l+1)-U(l)If the | | < the value, the operation is terminated, otherwise, the steps 3 to 6 are continuously circulated until the condition is met.
The data after dimensionality reduction is analyzed by using improved GK clustering, and higher speed and higher precision are obtained.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.

Claims (3)

1. A soy sauce classification method based on a self-made electronic nose system is characterized by comprising the following steps:
s1, opening the bottle cover of the sample, and electrifying the electronic nose for 10 minutes for preheating;
s2, putting a 15 ml soy sauce sample into a sample bottle, connecting an air pipe of the sample bottle with an air chamber, placing the sample bottle in the center of the air chamber, and standing the sensor for 60 minutes until the gas is completely volatilized;
s3, collecting and recording data by using an upper computer program compiled by NI LabVIEW 2018, collecting the data once every 5 minutes for three times in total, and taking the average value of the three collection results as stable data of a soy sauce sample;
and S4, after the soy sauce sample is collected once, opening the box cover and standing for 10 minutes to enable each sensor to recover the initial state, and then repeatedly collecting other soy sauce samples. The state of the air chamber can be observed through a LabVIEW response curve, and after the air chamber is completely emptied, data of another soy sauce sample is collected;
steps S2-S4 are repeated until all data acquisitions are complete.
2. The soy sauce classification method based on the homemade electronic nose system according to claim 1, characterized in that: the electronic nose system for soy sauce mainly comprises a gas chamber (where a sensor array is arranged), a sample bottle, a gas sensor array, a data acquisition card, a computer and the like.
3. The soy sauce classification method based on the homemade electronic nose system according to claim 2, characterized in that: the gas sensor for selecting the self-made electronic nose aiming at the characteristic aroma of the soy sauce is required to be sensitive to the odor components and the concentration of a soy sauce sample, and eight semiconductor gas sensors of MQ135, MQ3, TGS2600, TGS2620, TGS2602, TGS2610, TGS2611 and TGS813 are selected for sample identification.
CN202010516889.4A 2020-06-09 2020-06-09 Soy sauce classification method based on self-made electronic nose system Withdrawn CN111638246A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010516889.4A CN111638246A (en) 2020-06-09 2020-06-09 Soy sauce classification method based on self-made electronic nose system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010516889.4A CN111638246A (en) 2020-06-09 2020-06-09 Soy sauce classification method based on self-made electronic nose system

Publications (1)

Publication Number Publication Date
CN111638246A true CN111638246A (en) 2020-09-08

Family

ID=72330007

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010516889.4A Withdrawn CN111638246A (en) 2020-06-09 2020-06-09 Soy sauce classification method based on self-made electronic nose system

Country Status (1)

Country Link
CN (1) CN111638246A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113567516B (en) * 2021-06-28 2023-05-16 滁州职业技术学院 Sulfamethoxypyrimidine molecularly imprinted electrode and preparation method and application thereof

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113567516B (en) * 2021-06-28 2023-05-16 滁州职业技术学院 Sulfamethoxypyrimidine molecularly imprinted electrode and preparation method and application thereof

Similar Documents

Publication Publication Date Title
JP2022137166A (en) System and method of classification of biological particles
Lewicki A review of methods for spike sorting: the detection and classification of neural action potentials
CN109002859B (en) Sensor array feature selection and array optimization method based on principal component analysis
CN112699793A (en) Fatigue driving detection optimization identification method based on random forest
CN110378374B (en) Tea near infrared spectrum classification method for extracting fuzzy identification information
CN105954412A (en) Sensor array optimization method for Carya cathayensis freshness detection
Mavi et al. Mango ripeness classification system using hybrid technique
Jiang et al. Identification of tomato maturity based on multinomial logistic regression with kernel clustering by integrating color moments and physicochemical indices
Bittencourt et al. Feature selection by using classification and regression trees (CART)
CN112244871A (en) Amplitude integration electroencephalogram classification recognition system based on machine learning
CN111638246A (en) Soy sauce classification method based on self-made electronic nose system
CN112232387A (en) LSELM-RFE-based effective characteristic identification method for disease symptoms of grain crops
CN114707608A (en) Medical quality control data processing method, apparatus, device, medium, and program product
CN102749420B (en) Peanut quality determination and evaluation methods suitable for gel type protein processing
CN106940292A (en) Bar denier wood raw material quick nondestructive discrimination method of damaging by worms based on multi-optical spectrum imaging technology
CN104990891B (en) A kind of seed near infrared spectrum and spectrum picture qualitative analysis model method for building up
CN110929888A (en) Grain crop disease prediction method based on random compact Gaussian kernel function
CN107121408A (en) The quick nondestructive discrimination method of edible vegetable oil kind
Geollegue et al. Seed of rice plant classification using coarse tree classifier
CN108399433A (en) A kind of sorting technique based on Dactylogram Chart About Chinese Spirit feature
CN106979998A (en) The bionic olfactory Rapid non-destructive testing device and detection method of apple freshness
CN112036482A (en) Traditional Chinese medicine classification method based on electronic nose sensor data
CN109685099B (en) Apple variety distinguishing method based on spectrum band optimization fuzzy clustering
CN113933334B (en) Locust honey authenticity identification method based on feature selection and machine learning algorithm
CN206710409U (en) The bionic olfactory Rapid non-destructive testing device of apple freshness

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20200908