CN105954451B - Cigarette type Quick method based on the panchromatic modal data of electronic nose - Google Patents
Cigarette type Quick method based on the panchromatic modal data of electronic nose Download PDFInfo
- Publication number
- CN105954451B CN105954451B CN201610398897.7A CN201610398897A CN105954451B CN 105954451 B CN105954451 B CN 105954451B CN 201610398897 A CN201610398897 A CN 201610398897A CN 105954451 B CN105954451 B CN 105954451B
- Authority
- CN
- China
- Prior art keywords
- odor type
- cigarette
- brand cigarettes
- panchromatic
- brand
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Manufacturing Of Cigar And Cigarette Tobacco (AREA)
Abstract
The present invention relates to a kind of cigarette type Quick method based on the panchromatic modal data of electronic nose, this method is differentiated using the fine difference between the panchromatic modal data of different type cigarette by building supporting vector machine model to make a distinction.It is demonstrated experimentally that the accuracy rate of the differentiation result of method provided by the invention is higher, the requirement that cigarette differentiates disclosure satisfy that.
Description
Technical field
The present invention relates to tobacco detection field, more particularly, to a kind of cigarette class based on the panchromatic modal data of electronic nose
Type Quick method.
Background technology
The research of electronic nose starts from the phase early 1980s, and it passes through the volatility gas to having in object
Body, organic matter etc. are trapped, and then carry out statistical analysis to the data trapped.It is imitated using a series of sensor
Smell, the smell of complex sample can be detected and distinguished, there is the advantages of cost is cheap, is widely used.Nearly ten years, gone out
A large amount of reports that every research is carried out using electronic nose are showed, the application being included in tobacco quality evaluation.Cut-off 2008, entirely
The supplier of world's electronic nose commercial product has reached 18, such as French Alpha MOS, the Cyrano sciences in the U.S.
Deng.Most of electronic noses are sensor type, and unfortunately, intoxicating phenomenon easily occurs for the electronic nose of sensor type.
The HERACLES Flash GC type electronic noses of French Alpha MOS productions, its built-in Trap, can be greatly improved
Detection sensitivity；Using post sheath heating technique, heating rate reaches as high as 25 DEG C/s；Using the principle of gas-chromatography, configuration two
The different chromatographic column of root polarity and two fid detectors carry out gathered data, and its post footpath is 0.1mm, have high theoretical tray
Number.
The content of the invention
The present invention is the problem for solving above prior art, there is provided a kind of cigarette class based on the panchromatic modal data of electronic nose
Type Quick method, this method are supported using the fine difference between the panchromatic modal data of different type cigarette by building
Vector machine model differentiates to make a distinction.It is demonstrated experimentally that the accuracy rate that method provided by the invention differentiates is higher, volume disclosure satisfy that
The requirement that cigarette differentiates.
To realize above goal of the invention, the technical scheme of use is：
A kind of cigarette type Quick method based on the panchromatic modal data of electronic nose, for differentiating whether cigarette belongs to certain
One brand, then on the basis of differentiating that cigarette belongs to a certain brand, the valency class belonging to cigarette and odor type are differentiated, it is described
Method of discrimination comprises the following steps：
First, brand judges the stage
S11. the cigarette of a certain brand is referred to as A brand cigarettes, gathers A brand cigarettes using electronic nose, other brands are rolled up
The panchromatic modal data of cigarette, and the panchromatic modal data of collection is equally divided into k parts, it is training data to make (k-1) part, and 1 part is test
Data；
S12. supporting vector machine model is built：
In formula (1), yiRepresent sample class, sample class includes positive sample and negative sample, wherein the value of positive sample be+
1, the value of negative sample is -1；K(xiX) kernel function is represented, m represents total sample number, xiIt is the vector for representing sample, x represents all
The average vector of sample, α* iAnd b*For model parameter, sgn is output function；
S13. and then (k-1) part training data is used to be trained successively to supporting vector machine model, the mistake being trained
The panchromatic modal data of Cheng Zhong, A brand cigarette is as positive sample, and the panchromatic modal data of other brand cigarettes is as negative sample；
S14. the cigarette brand belonging to test data is differentiated using the supporting vector machine model trained, judges it
Whether belong to A brand cigarettes, and export differentiation result；
S15. step S13, S14 repeats k times, in k implementation procedure, the conduct successively of the panchromatic modal data of k parts
Test data, and remaining panchromatic modal data is then used as training data；
S16. based on the k accuracy rate differentiated, the parameter of supporting vector machine model is optimized；
S17. the panchromatic modal data of cigarette to be discriminated is inputted, is then judged using the supporting vector machine model by optimization
Whether it belongs to A brands：If the output valve of supporting vector machine model is+1, represents that cigarette to be discriminated belongs to A brands, then perform
Step S21 and step S4；If the output valve of supporting vector machine model is -1, then it represents that cigarette to be discriminated belongs to other brands,
Now perform step S4；
2nd, to the differentiation of the affiliated valency class of cigarette to be discriminated
S21. A brand cigarettes are set and include m kind valency classes, respectively numbering be valency class 1, valency class 2, valency class 3 ..., valency class m, make
Gather the panchromatic modal data of the A brand cigarettes of various valency classes respectively with electronic nose；
S22. built and supported according to step S11~S16 method using the panchromatic modal data of the A brand cigarettes of m kind valency classes
Vector machine model, wherein with valency class i A during according to supporting vector machine model training of the step S13 method to structure
The panchromatic modal data of brand cigarette is positive sample, and the panchromatic modal data of the A brand cigarettes of its partial valence class is negative sample, and i's is initial
It is worth for 1；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If SVMs mould
Type output+1, then it represents that cigarette to be discriminated belongs to valency class i A brand cigarettes, now performs step S31 and step S4；If support
Vector machine model output -1, then it represents that cigarette to be discriminated is not belonging to valency class i A brand cigarettes, now performs step S23；
S23. make i=i+1 and repeat step S22；
3rd, to the differentiation of the affiliated odor type of cigarette to be discriminated
S31. A brand cigarettes are set and include n kind odor types, respectively numbering be odor type 1, odor type 2, odor type 3 ... odor type n, use
Electronic nose gathers the panchromatic modal data of the A brand cigarettes of various odor types respectively；
S32. built and supported according to step S11~S16 method using the panchromatic modal data of the A brand cigarettes of n kind odor types
Vector machine model, wherein with odor type p A during according to supporting vector machine model training of the step S13 method to structure
The panchromatic modal data of brand cigarette is positive sample, and the panchromatic modal data of the A brand cigarettes of remaining odor type is negative sample, and p's is initial
It is worth for 1；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If SVMs mould
Type output+1, then it represents that cigarette to be discriminated belongs to odor type p A brand cigarettes, performs step S4；If supporting vector machine model is defeated
Go out -1, then it represents that cigarette to be discriminated is not belonging to odor type p A brand cigarettes, now performs step S23；
S33. make p=p+1 and repeat step S32；
S4. output differentiates result.
In such scheme, by build supporting vector machine model and using the panchromatic modal data of cigarette being trained and
Test so that the parameter of supporting vector machine model can with various brand cigarettes, valency class cigarette, odor type cigarette panchromatic modal data
Corresponding contact is established, closer in the actual conditions of cigarette, therefore the accuracy rate of the differentiation result of method provided by the invention
It is higher, it disclosure satisfy that the requirement that cigarette differentiates.
Preferably, the A brand cigarettes include three kinds of valency classes, are respectively a kind of, two classes and three classes, then to volume to be discriminated
The process that the affiliated valency class of cigarette differentiates is specific as follows：
S101. using electronic nose gather respectively a kind of A brand cigarettes, two class A brand cigarettes, three class A brand cigarettes it is complete
Chromatographic data；
S102. using a kind of A brand cigarettes, two class A brand cigarettes, the panchromatic modal data of three class A brand cigarettes according to step
Rapid S11~S16 method structure supporting vector machine model, wherein according to step S13 method to the SVMs to structure
Using the panchromatic modal data of a kind of A brand cigarettes as positive sample during model training, two class A brand cigarettes, three class A brands volume
The panchromatic modal data of cigarette is negative sample；Then the whole chromatogram number of cigarette to be discriminated is inputted to the supporting vector machine model after optimization
According to；If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to a kind of A brand cigarettes, now perform step S31 and
Step S4；If supporting vector machine model output -1, then it represents that cigarette to be discriminated belongs to two class A brand cigarettes or three class A brands volume
Cigarette, now perform step S103；
S103. using a kind of A brand cigarettes, two class A brand cigarettes, the panchromatic modal data of three class A brand cigarettes according to step
Rapid S11~S16 method structure supporting vector machine model, wherein according to step S13 method to the SVMs to structure
Using the panchromatic modal data of two class A brand cigarettes as positive sample during model training, a kind of A brand cigarettes, three class A brands volume
The panchromatic modal data of cigarette is negative sample；Then the whole chromatogram number of cigarette to be discriminated is inputted to the supporting vector machine model after optimization
According to；If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to two class A brand cigarettes, now perform step S31 and
Step S4；If supporting vector machine model output -1, then it represents that cigarette to be discriminated belongs to three class A brand cigarettes, now performs step
S104；
S104. using a kind of A brand cigarettes, two class A brand cigarettes, the panchromatic modal data of three class A brand cigarettes according to step
Rapid S11~S16 method structure supporting vector machine model, wherein according to step S13 method to the SVMs to structure
Using the panchromatic modal data of three class A brand cigarettes as positive sample during model training, a kind of A brand cigarettes, two class A brands volume
The panchromatic modal data of cigarette is negative sample；Then the whole chromatogram number of cigarette to be discriminated is inputted to the supporting vector machine model after optimization
According to；If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to three class A brand cigarettes, now perform step S31 and
Step S4.
Preferably, A brand cigarettes share five kinds of odor types, and respectively I odor type, II odor type, III odor type, IV odor type and V are fragrant
Type, the then detailed process differentiated to the affiliated odor type of cigarette to be discriminated are as follows：
S201. the panchromatic modal data for gathering five kinds of odor type A brand cigarettes respectively is gathered respectively using electronic nose；
S202. rolled up using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brands
Cigarette, the panchromatic modal data of V odor type A brand cigarettes build supporting vector machine model according to step S11~S16 method, wherein pressing
According to step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram number of I odor type A brand cigarettes
According to for positive sample, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, V odor type A brand cigarettes
Panchromatic modal data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；
If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to I odor type, now performs step S4；If SVMs
Model output -1, then it represents that cigarette to be discriminated belongs to II odor type, III odor type, IV odor type or V odor type, now performs step S203；
S203. rolled up using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brands
Cigarette, the panchromatic modal data of V odor type A brand cigarettes build supporting vector machine model according to step S11~S16 method, wherein pressing
According to step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram of II odor type A brand cigarettes
Data are positive sample, I odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, V odor type A brand cigarettes
Panchromatic modal data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；
If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to II odor type, now performs step S4；If SVMs
Model output -1, then it represents that cigarette to be discriminated belongs to III odor type, IV odor type or V odor type, now performs step S204；
S204. rolled up using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brands
Cigarette, the panchromatic modal data of V odor type A brand cigarettes build supporting vector machine model according to step S11~S16 method, wherein pressing
According to step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram of III odor type A brand cigarettes
Data are positive sample, I odor type A brand cigarettes, II odor type A brand cigarettes, IV odor type A brand cigarettes, V odor type A brand cigarettes
Panchromatic modal data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；
If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to III odor type, now performs step S4；If SVMs
Model output -1, then it represents that cigarette to be discriminated belongs to IV odor type or V odor type, now performs step S205；
S205. rolled up using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brands
Cigarette, the panchromatic modal data of V odor type A brand cigarettes build supporting vector machine model according to step S11~S16 method, wherein pressing
According to step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram of IV odor type A brand cigarettes
Data are positive sample, I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, V odor type A brand cigarettes
Panchromatic modal data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；
If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to IV odor type, now performs step S4；If SVMs
Model output -1, then it represents that cigarette to be discriminated belongs to V odor type, now performs step S206；
S206. rolled up using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brands
Cigarette, the panchromatic modal data of V odor type A brand cigarettes build supporting vector machine model according to step S11~S16 method, wherein pressing
According to step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram of V odor type A brand cigarettes
Data are positive sample, I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes
Panchromatic modal data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；
If supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to V odor type, now performs step S4.
Preferably, in described step S11, S21, S31, it is necessary to enter to panchromatic modal data after panchromatic modal data is collected
Row normalized.
Preferably, the procedural representation of the normalized is as follows：
Wherein xi(j) j-th of chromatographic data value of cigarette type i panchromatic modal data is represented, the cigarette type includes
Brand, valency class or odor type, x (j)minWith x (j)maxJ-th chromatographic data is represented in cigarette type i panchromatic modal data respectively
Minimum value and maximum.
Preferably, the supporting vector machine model is run in Matlab environment.
Preferably, described step S11, S21, S31 gather the panchromatic modal data of cigarette using Flash GC types electronic nose.It is excellent
Selection of land, for the Flash GC types electronic nose in the panchromatic modal data of collection cigarette, it sets condition as follows：Initial temperature：45℃；Eventually
Temperature：270℃；Programming rate：2℃·s-1；Sample introduction needle scavenging period：90s；Injector temperature：210℃；FID temperature：260℃；
TRAP initial temperature：40℃；TRAP final temperatures：260℃；Sample introduction needle temperature：110℃；Incubation temperature：90℃；Brooding time：1200s；Catch
Collect the time：50s；Sample size：500μL.
Compared with prior art, the beneficial effects of the invention are as follows：
Method provided by the invention is using the fine difference between the panchromatic modal data of different type cigarette, by building branch
Hold vector machine model and differentiate to make a distinction.It is demonstrated experimentally that the accuracy rate of the differentiation result of method provided by the invention is higher, energy
Enough meet the requirement that cigarette differentiates.
Brief description of the drawings
Fig. 1 is the flow chart of cigarette method of discrimination.
Fig. 2 is ROC and PRC curve maps.
Fig. 3 is the visual effect figure that A brand cigarettes differentiate with non-A brand cigarettes.
Fig. 4 is the differentiation procedure chart to A brand cigarette odor types.
Fig. 5 is the visual effect figure differentiated to A brand cigarettes odor type.
Fig. 6 is the flow chart differentiated to each valency class cigarette of A brand boards.
The visual effect figure that Fig. 7 differentiates to the class cigarette of I odor type A brands three.
The visual effect figure that Fig. 8 is a kind of to III odor type and three class A brand cigarettes are differentiated.
Fig. 9 is the visual effect figure differentiated to IV two kinds of cigarette of odor type A16 and A19.
The visual effect figure that Figure 10 is a kind of to V odor type and two class A brand cigarettes are differentiated.
Embodiment
Accompanying drawing being given for example only property explanation, it is impossible to be interpreted as the limitation to this patent；
Below in conjunction with drawings and examples, the present invention is further elaborated.
Embodiment 1
Cigarette type Quick method provided by the invention is used to differentiate whether cigarette belongs to a certain brand, is then sentencing
Other cigarette belongs on the basis of a certain brand, and the valency class belonging to cigarette and odor type are differentiated, comprised the following steps：
First, brand judges the stage
S11. the cigarette of a certain brand is referred to as A brand cigarettes, gathers A brand cigarettes using electronic nose, other brands are rolled up
The panchromatic modal data of cigarette, and the panchromatic modal data of collection is equally divided into k parts, it is training data to make (k-1) part, and 1 part is test
Data；
S12. supporting vector machine model is built：
In formula (1), yiRepresent sample class, sample class includes positive sample and negative sample, wherein the value of positive sample be+
1, the value of negative sample is -1；K(xiX) kernel function is represented, m represents total sample number, xiIt is the vector for representing sample, x represents all
The average vector of sample, α* iAnd b*For model parameter, sgn is output function；
S13. and then (k-1) part training data is used to be trained successively to supporting vector machine model, the mistake being trained
The panchromatic modal data of Cheng Zhong, A brand cigarette is as positive sample, and the panchromatic modal data of other brand cigarettes is as negative sample；
S14. the cigarette brand belonging to test data is differentiated using the supporting vector machine model trained, judges it
Whether belong to A brand cigarettes, and export differentiation result；
S15. step S13, S14 repeats k times, in k implementation procedure, the conduct successively of the panchromatic modal data of k parts
Test data, and remaining panchromatic modal data is then used as training data；
S16. based on the k accuracy rate differentiated, the parameter of supporting vector machine model is optimized；
S17. the panchromatic modal data of cigarette to be discriminated is inputted, is then judged using the supporting vector machine model by optimization
Whether it belongs to A brands：If the output valve of supporting vector machine model is+1, represents that cigarette to be discriminated belongs to A brands, then perform
Step S21 and step S4；If the output valve of supporting vector machine model is -1, then it represents that cigarette to be discriminated belongs to other brands,
Now perform step S4；
2nd, to the differentiation of the affiliated valency class of cigarette to be discriminated
S21. A brand cigarettes are set and include m kind valency classes, respectively numbering be valency class 1, valency class 2, valency class 3 ..., valency class m, make
Gather the panchromatic modal data of the A brand cigarettes of various valency classes respectively with electronic nose；
S22. built and supported according to step S11~S16 method using the panchromatic modal data of the A brand cigarettes of m kind valency classes
Vector machine model, wherein with valency class i A during according to supporting vector machine model training of the step S13 method to structure
The panchromatic modal data of brand cigarette is positive sample, and the panchromatic modal data of the A brand cigarettes of its partial valence class is negative sample, and i's is initial
It is worth for 1；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If SVMs mould
Type output+1, then it represents that cigarette to be discriminated belongs to valency class i A brand cigarettes, now performs step S31 and step S4；If support
Vector machine model output -1, then it represents that cigarette to be discriminated is not belonging to valency class i A brand cigarettes, now performs step S23；
S23. make i=i+1 and repeat step S22；
3rd, to the differentiation of the affiliated odor type of cigarette to be discriminated
S31. A brand cigarettes are set and include n kind odor types, respectively numbering be odor type 1, odor type 2, odor type 3 ... odor type n, use
Electronic nose gathers the panchromatic modal data of the A brand cigarettes of various odor types respectively；
S32. built and supported according to step S11~S16 method using the panchromatic modal data of the A brand cigarettes of n kind odor types
Vector machine model, wherein with odor type p A during according to supporting vector machine model training of the step S13 method to structure
The panchromatic modal data of brand cigarette is positive sample, and the panchromatic modal data of the A brand cigarettes of remaining odor type is negative sample, and p's is initial
It is worth for 1；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If SVMs mould
Type output+1, then it represents that cigarette to be discriminated belongs to odor type p A brand cigarettes, performs step S4；If supporting vector machine model is defeated
Go out -1, then it represents that cigarette to be discriminated is not belonging to odor type p A brand cigarettes, now performs step S23；
S33. make p=p+1 and repeat step S32；
S4. output differentiates result.
Embodiment 2
On the basis of embodiment 1, the cigarette of selected a collection of different brands, valency class and odor type has been carried out specifically the present embodiment
Experiment.The cigarette brand that this experiment uses includes A brands, B brands, C brands, D brands, E brands, F brands and G product
Board.And when carrying out valency class, odor type judging, mainly use different price class, the cigarette of odor type of A brands, A brand cigarette bags
Three kinds of valency classes have been included, have respectively been divided into a kind of, two classes and three classes, in the present embodiment, one kind, two classes, three class A brand cigarettes are distinguished
It is expressed as A1, A2, A3；A brand cigarettes include five kinds of odor types, respectively I odor type, II odor type, III odor type, IV odor type and V
Odor type.This experiment is to A brands and non-A brand cigarettes, the cigarette of the various odor types of A brands, the cigarette of the various valency classes of A brands, I perfume
Type A brands one kind, two classes and three class cigarette, II odor type A brand cigarettes, III odor type one kind and three class A brand cigarettes, IV odor type
Two kinds of cigarette of A16 and A19, V odor type are a kind of and two class A brand cigarettes carry out specific discriminating experiment.
This experiment gathers the panchromatic modal data of cigarette using Flash GC types electronic nose, and Flash GC type electronic noses are set
It is as follows to put condition：Initial temperature：45℃；Final temperature：270℃；Programming rate：2℃·s-1；Sample introduction needle scavenging period：90s；Injection port temperature
Degree：210℃；FID temperature：260℃；TRAP initial temperature：40℃；TRAP final temperatures：260℃；Sample introduction needle temperature：110℃；Hatching temperature
Degree：90℃；Brooding time：1200s；Trap time：50s；Sample size：500μL.
It is below the concrete condition of experiment：
(1) differentiation of A brands and non-A brand cigarettes
The parameter such as institutes of table 1 such as the accuracy that structure supporting vector machine model is differentiated to A brands-non-A brand cigarettes
Show, obtain 93.28% accuracy, 96.18% sensitiveness, 83.64% specificity, 95.12% accuracy rate and
0.8089 geneva coefficient correlation.ROC and PRC curves are as shown in Fig. 2 area is respectively 0.9865 and 0.9952.As a result show,
The supporting vector machine model established can differentiate A brands and non-A brand cigarettes well.A brand cigarettes and non-A brand cigarettes
The visual effect of differentiation is as shown in Figure 3.
The accuracy rate that the A brands of table 1-non-A brand cigarettes differentiate
(2) to the differentiation of the various odor type cigarette of A brands
It is specific as shown in Figure 4 to the differentiation process of A brand cigarette odor types.Supporting vector machine model is built to A brand cigarettes
Ith, the differentiation accuracy of II, III, IV, V odor type is as shown in table 2.Differentiating accuracy is respectively：92.95%th, 89.78%,
100%th, 57.50% and 71.67%.Show that supporting vector machine model can differentiate the A brands of I, II, III, V odor type well
Cigarette.Wherein IV odor type A brand cigarette numbers of samples are 120, wherein 51 samples, which are supported vector machine model, is determined as V perfume
Type A brand cigarettes.V odor type A brand cigarettes number of samples is 180, wherein 51 samples are identified as IV odor type A brand cigarettes.
This is due to IV odor type and the reason for V odor type odor type is closer to, and causes chromatographic data more similar, it is desirable to which this two class is complete
It is complete to separate more difficulty.The visual effect differentiated based on A brand cigarettes odor type is as shown in Figure 5.
The A brand cigarettes odor type of table 2 differentiates result
(3) to the differentiation of the various valency class cigarette of A brands
1) to the differentiation of the various valency class cigarette of A brands
The present embodiment is classified research based on the flow chart shown in Fig. 6 to each valency class of A brand cigarettes.Such as the institute of table 3
Show, supporting vector machine model is to A brand cigarettes one kind and two classes, three classes, two classes and a kind of, three classes and three classes and one kind, three
The differentiation accuracy of class price is 100%, shows that supporting vector machine model can very well differentiate the A brands of different price class
Cigarette.
The different price class of table 3 differentiates result
2) differentiation of, two classes a kind of to A brands and three class cigarette
The present embodiment has carried out specific differentiation to each trade mark cigarette in A brand one kind cigarette, differentiates result such as table 4
It is shown.A11, A12, A13, A14, A15, A16, A17, A18, A19 are a kind of cigarette, and its degree of accuracy differentiated is 100%,
Show that method provided by the invention can differentiate a kind of A brand cigarettes well.
A kind of A brand cigarettes of table 4 differentiate result
The present embodiment differentiated using supporting vector machine model to each trade mark cigarette of the class cigarette of A brands two,
A21, A22 and A23 are the class cigarette of A brands two, and it differentiates that result is as shown in table 5.A21, A22 and A23 the differentiation degree of accuracy are all
It is 100%, shows that the supporting vector machine model of structure can be good at distinguishing each trade mark cigarette of the class cigarette of A brands two.
The class A brand cigarettes of table 5 two differentiate result
The present embodiment differentiated using supporting vector machine model to each trade mark cigarette of the class cigarette of A brands three,
A31, A32, A34, A35, A36, A37, A38, A39, A310 are the class cigarette of A brands three, differentiate the confusion matrix such as table 6 of result
It is shown.A31, A32, A34, A35, A36, A37, A38, A39, A310 the differentiation degree of accuracy are 100% respectively, and A33 differentiation
The degree of accuracy is 98.33%.The method for showing to propose can be in the extraordinary class cigarette of discrimination A brands three each trade mark cigarette.
The differentiation result of the class A brand cigarettes of table 6 three
| A31/ sample numbers | 60 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| A32/ sample numbers | 0 | 60 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| A33/ sample numbers | 0 | 1 | 59 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| A34/ sample numbers | 0 | 0 | 0 | 60 | 0 | 0 | 0 | 0 | 0 | 0 |
| A35/ sample numbers | 0 | 0 | 0 | 0 | 60 | 0 | 0 | 0 | 0 | 0 |
| A36/ sample numbers | 0 | 0 | 0 | 0 | 0 | 60 | 0 | 0 | 0 | 0 |
| A37/ sample numbers | 0 | 0 | 0 | 0 | 0 | 0 | 30 | 0 | 0 | 0 |
| A38/ sample numbers | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 30 | 0 | 0 |
| A39/ sample numbers | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 60 | 0 |
| A310/ sample numbers | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 30 |
| A31 | A32 | A33 | A34 | A35 | A36 | A37 | A38 | A39 | A310 |
3) to the differentiation of I odor type A brand cigarettes
Supporting vector machine model provided by the invention is a kind of to I odor type A brands, the differentiation accuracy of two classes and three classes is all
100%, the method for showing to propose can be good at distinguishing the class A brand cigarettes of I odor type three.It is and pre- to the class cigarette of I odor type A brands three
The visual effect of survey is as shown in Figure 7.
4) to the differentiation of II odor type A brand cigarettes
The present invention has also carried out differentiating research to II odor type A brand cigarettes, and A31, A34, A35 and A36 are II odor type A product
Board cigarette, differentiate that the confusion matrix of result is as shown in table 7.Differentiation of the supporting vector machine model to A31, A34, A35 and A36 is accurate
Rate is 100%.For A33, only one sample is erroneously identified as A34, precision of prediction 98.33%.These results indicate that
Supporting vector machine model can be good at identifying II odor type A brand cigarettes.
The differentiation result of the odor type A brand cigarettes of table 7 II
| A31/ sample numbers | 60 | 0 | 0 | 0 | 0 |
| A33/ sample numbers | 0 | 59 | 1 | 0 | 0 |
| A34/ sample numbers | 0 | 0 | 60 | 0 | 0 |
| A35/ sample numbers | 0 | 0 | 0 | 60 | 0 |
| A36/ sample numbers | 0 | 0 | 0 | 0 | 60 |
| A31 | A33 | A34 | A35 | A36 |
5) a kind of to the III odor type and differentiation of three class A brand cigarettes
In the present embodiment, also a kind of to III odor type and three class A brand cigarettes have carried out differentiating research, and its visual effect figure is such as
Shown in Fig. 8, and differentiate that result is as shown in table 8.Supporting vector machine model is a kind of to III odor type and the prediction of three class A brand cigarettes is smart
Degree, Sensitivity and Specificity etc. are all 100.0%, and geneva coefficient correlation is 1.00.Show that supporting vector machine model can be accurate
Ground identifies III odor type A brands one kind and three class cigarette.
The odor type of table 8 III is a kind of and three class A brand cigarettes differentiate result
6) to the differentiation of IV odor type two kinds of cigarette of A16 and A19
In the present embodiment, supporting vector machine model has carried out differentiating research to IV two kinds of cigarette of odor type A16 and A19, and it is sentenced
The visual effect of other result is as shown in Figure 9.As can be seen that supporting vector machine model can accurately identify IV odor type A16 and A19
Two class brand cigarettes, indicate the validity of current method.
7) a kind of to the V odor type and differentiation of two class A brand cigarettes
In the present embodiment, supporting vector machine model is a kind of to V odor type and two class A brand cigarettes have carried out differentiating research, its
Differentiate that the visual effect of result is as shown in Figure 10.As can be seen that supporting vector machine model can accurately identify V odor type one kind
With two class cigarette, the validity of current method is indicated.
Differentiation result more than in each form and figure can be seen that using panchromatic modal data can obtain it is higher
Prediction result, because each type cigarette is respectively provided with the characteristic fingerprint chromatogram of oneself.Chromatographic peak area is used with other
Method compare, the fine difference of chromatographic data between different type cigarette can be caught using panchromatic modal data, so can
Produce higher differentiation result.
Obviously, the above embodiment of the present invention is only intended to clearly illustrate example of the present invention, and is not pair
The restriction of embodiments of the present invention.For those of ordinary skill in the field, may be used also on the basis of the above description
To make other changes in different forms.There is no necessity and possibility to exhaust all the enbodiments.It is all this
All any modification, equivalent and improvement made within the spirit and principle of invention etc., should be included in the claims in the present invention
Protection domain within.
Claims (8)
- A kind of 1. cigarette type Quick method based on the panchromatic modal data of electronic nose, for differentiating whether cigarette belongs to a certain Brand, then on the basis of differentiating that cigarette belongs to a certain brand, the valency class belonging to cigarette and odor type are differentiated, its feature It is：The method of discrimination comprises the following steps：First, brand judges the stageS11. the cigarette of a certain brand is referred to as A brand cigarettes, A brand cigarettes, other brand cigarettes is gathered using electronic nose Panchromatic modal data, and the panchromatic modal data of collection is equally divided into k parts, it is training data to make (k-1) part, and 1 part is test data；S12. supporting vector machine model is built：<mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>sgn</mi> <mo>{</mo> <msubsup> <mi>&Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>m</mi> </msubsup> <msubsup> <mi>&alpha;</mi> <mi>i</mi> <mo>*</mo> </msubsup> <msub> <mi>y</mi> <mi>i</mi> </msub> <mi>K</mi> <mrow> <mo>(</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>&CenterDot;</mo> <mi>x</mi> <mo>)</mo> </mrow> <mo>+</mo> <msup> <mi>b</mi> <mo>*</mo> </msup> <mo>}</mo> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>In formula (1), yiSample class is represented, sample class includes positive sample and negative sample, and the wherein value of positive sample is+1, is born The value of sample is -1；K(xiX) kernel function is represented, m represents total sample number, xiIt is the vector for representing sample, x represents all samples Average vector, α* iAnd b*For model parameter, sgn is output function；S13. and then (k-1) part training data is used to be trained successively to supporting vector machine model, during being trained, The panchromatic modal data of A brand cigarettes is as positive sample, and the panchromatic modal data of other brand cigarettes is as negative sample；S14. using the supporting vector machine model trained to the cigarette brand belonging to test data differentiate whether judge it Belong to A brand cigarettes, and export differentiation result；S15. step S13, S14 repeats k times, and in k implementation procedure, the panchromatic modal data of k parts is successively as test Data, and remaining panchromatic modal data is then used as training data；S16. based on the k accuracy rate differentiated, the parameter of supporting vector machine model is optimized；S17. the panchromatic modal data of cigarette to be discriminated is inputted, then judges that it is using the supporting vector machine model by optimization It is no to belong to A brands：If the output valve of supporting vector machine model is+1, represents that cigarette to be discriminated belongs to A brands, then perform step S21 and step S4；If the output valve of supporting vector machine model is -1, then it represents that cigarette to be discriminated belongs to other brands, now Perform step S4；2nd, to the differentiation of the affiliated valency class of cigarette to be discriminatedS21. A brand cigarettes are set and include m kind valency classes, respectively numbering be valency class 1, valency class 2, valency class 3 ..., valency class m, use electricity Sub- nose gathers the panchromatic modal data of the A brand cigarettes of various valency classes respectively；S22. supporting vector is built according to step S11~S16 method using the panchromatic modal data of the A brand cigarettes of m kind valency classes Machine model, wherein with valency class i A brands during according to supporting vector machine model training of the step S13 method to structure The panchromatic modal data of cigarette is positive sample, and the panchromatic modal data of the A brand cigarettes of its partial valence class is negative sample, and i initial value is 1；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If supporting vector machine model is defeated Go out+1, then it represents that cigarette to be discriminated belongs to valency class i A brand cigarettes, now performs step S31 and step S4；If supporting vector Machine model output -1, then it represents that cigarette to be discriminated is not belonging to valency class i A brand cigarettes, now performs step S23；S23. make i=i+1 and repeat step S22；3rd, to the differentiation of the affiliated odor type of cigarette to be discriminatedS31. A brand cigarettes are set and include n kind odor types, respectively numbering be odor type 1, odor type 2, odor type 3 ... odor type n, use electronics Nose gathers the panchromatic modal data of the A brand cigarettes of various odor types respectively；S32. supporting vector is built according to step S11~S16 method using the panchromatic modal data of the A brand cigarettes of n kind odor types Machine model, wherein with odor type p A brands during according to supporting vector machine model training of the step S13 method to structure The panchromatic modal data of cigarette is positive sample, and the panchromatic modal data of the A brand cigarettes of remaining odor type is negative sample, and p initial value is 1；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If supporting vector machine model is defeated Go out+1, then it represents that cigarette to be discriminated belongs to odor type p A brand cigarettes, performs step S4；If supporting vector machine model output -1, Then represent that cigarette to be discriminated is not belonging to odor type p A brand cigarettes, now perform step S23；S33. make p=p+1 and repeat step S32；S4. output differentiates result.
- 2. the cigarette type Quick method according to claim 1 based on the panchromatic modal data of electronic nose, its feature exist In：The A brand cigarettes include three kinds of valency classes, are respectively a kind of, two classes and three classes, then the affiliated valency class of cigarette to be discriminated are sentenced Other process is specific as follows：S101. a kind of A brand cigarettes, two class A brand cigarettes, the whole chromatogram of three class A brand cigarettes are gathered respectively using electronic nose Data；S102. using a kind of A brand cigarettes, two class A brand cigarettes, the panchromatic modal data of three class A brand cigarettes according to step S11 ~S16 method structure supporting vector machine model, wherein according to step S13 method to the supporting vector machine model instruction to structure Using the panchromatic modal data of a kind of A brand cigarettes as positive sample in experienced process, two class A brand cigarettes, three class A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to a kind of A brand cigarettes, now performs step S31 and step S4；If supporting vector machine model output -1, then it represents that cigarette to be discriminated belongs to two class A brand cigarettes or three class A brand cigarettes, this Shi Zhihang steps S103；S103. using a kind of A brand cigarettes, two class A brand cigarettes, the panchromatic modal data of three class A brand cigarettes according to step S11 ~S16 method structure supporting vector machine model, wherein according to step S13 method to the supporting vector machine model instruction to structure Using the panchromatic modal data of two class A brand cigarettes as positive sample in experienced process, a kind of A brand cigarettes, three class A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to two class A brand cigarettes, now performs step S31 and step S4；If supporting vector machine model output -1, then it represents that cigarette to be discriminated belongs to three class A brand cigarettes, now performs step S104；S104. using a kind of A brand cigarettes, two class A brand cigarettes, the panchromatic modal data of three class A brand cigarettes according to step S11 ~S16 method structure supporting vector machine model, wherein according to step S13 method to the supporting vector machine model instruction to structure Using the panchromatic modal data of three class A brand cigarettes as positive sample in experienced process, a kind of A brand cigarettes, two class A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to three class A brand cigarettes, now performs step S31 and step S4。
- 3. the cigarette type Quick method according to claim 2 based on the panchromatic modal data of electronic nose, its feature exist In：A brand cigarettes share five kinds of odor types, respectively I odor type, II odor type, III odor type, IV odor type and V odor type, then to be discriminated The detailed process that the affiliated odor type of cigarette differentiates is as follows：S201. the panchromatic modal data for gathering five kinds of odor type A brand cigarettes respectively is gathered respectively using electronic nose；S202. using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, The panchromatic modal data of V odor type A brand cigarettes builds supporting vector machine model according to step S11~S16 method, wherein according to Step S13 method to during to the training of the supporting vector machine model of structure with the panchromatic modal data of I odor type A brand cigarettes For positive sample, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, V odor type A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to I odor type, now performs step S4；If SVMs mould Type output -1, then it represents that cigarette to be discriminated belongs to II odor type, III odor type, IV odor type or V odor type, now performs step S203；S203. using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, The panchromatic modal data of V odor type A brand cigarettes builds supporting vector machine model according to step S11~S16 method, wherein according to Step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram number of II odor type A brand cigarettes According to for positive sample, I odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, V odor type A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to II odor type, now performs step S4；If SVMs mould Type output -1, then it represents that cigarette to be discriminated belongs to III odor type, IV odor type or V odor type, now performs step S204；S204. using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, The panchromatic modal data of V odor type A brand cigarettes builds supporting vector machine model according to step S11~S16 method, wherein according to Step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram number of III odor type A brand cigarettes According to for positive sample, I odor type A brand cigarettes, II odor type A brand cigarettes, IV odor type A brand cigarettes, V odor type A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to III odor type, now performs step S4；If SVMs mould Type output -1, then it represents that cigarette to be discriminated belongs to IV odor type or V odor type, now performs step S205；S205. using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, The panchromatic modal data of V odor type A brand cigarettes builds supporting vector machine model according to step S11~S16 method, wherein according to Step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram number of IV odor type A brand cigarettes According to for positive sample, I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, V odor type A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to IV odor type, now performs step S4；If SVMs mould Type output -1, then it represents that cigarette to be discriminated belongs to V odor type, now performs step S206；S206. using I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes, The panchromatic modal data of V odor type A brand cigarettes builds supporting vector machine model according to step S11~S16 method, wherein according to Step S13 method to during to the training of the supporting vector machine model of structure with the whole chromatogram number of V odor type A brand cigarettes According to for positive sample, I odor type A brand cigarettes, II odor type A brand cigarettes, III odor type A brand cigarettes, IV odor type A brand cigarettes it is complete Chromatographic data is negative sample；Then the panchromatic modal data of cigarette to be discriminated is inputted to the supporting vector machine model after optimization；If Supporting vector machine model output+1, then it represents that cigarette to be discriminated belongs to V odor type, now performs step S4.
- 4. the cigarette type Quick method according to claim 1 based on the panchromatic modal data of electronic nose, its feature exist In：In described step S11, S21, S31, it is necessary to which panchromatic modal data is normalized after panchromatic modal data is collected.
- 5. the cigarette type Quick method according to claim 4 based on the panchromatic modal data of electronic nose, its feature exist In：The procedural representation of the normalized is as follows：<mrow> <msub> <mi>x</mi> <mi>i</mi> </msub> <msub> <mrow> <mo>(</mo> <mi>j</mi> <mo>)</mo> </mrow> <mrow> <mi>n</mi> <mi>e</mi> <mi>w</mi> </mrow> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>x</mi> <mi>i</mi> </msub> <mrow> <mo>(</mo> <mi>j</mi> <mo>)</mo> </mrow> <mo>-</mo> <mi>x</mi> <msub> <mrow> <mo>(</mo> <mi>j</mi> <mo>)</mo> </mrow> <mi>min</mi> </msub> </mrow> <mrow> <mi>x</mi> <msub> <mrow> <mo>(</mo> <mi>j</mi> <mo>)</mo> </mrow> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <mi>x</mi> <msub> <mrow> <mo>(</mo> <mi>j</mi> <mo>)</mo> </mrow> <mrow> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> </msub> </mrow> </mfrac> </mrow>Wherein xi(j) represent j-th of chromatographic data value of cigarette type i panchromatic modal data, the cigarette type include brand, Valency class or odor type, x (j)minWith x (j)maxThe minimum value of j-th of chromatographic data in cigarette type i panchromatic modal data is represented respectively And maximum.
- 6. the cigarette type Quick method according to claim 1 based on the panchromatic modal data of electronic nose, its feature exist In：The supporting vector machine model is run in Matlab environment.
- 7. the cigarette type Quick method based on the panchromatic modal data of electronic nose according to any one of claim 1~6, It is characterized in that：Described step S11, S21, S31 gather the panchromatic modal data of cigarette using Flash GC types electronic nose.
- 8. the cigarette type Quick method according to claim 7 based on the panchromatic modal data of electronic nose, its feature exist In：For the Flash GC types electronic nose in the panchromatic modal data of collection cigarette, it sets condition as follows：Initial temperature：45℃；Final temperature： 270℃；Programming rate：2℃·s-1；Sample introduction needle scavenging period：90s；Injector temperature：210℃；FID temperature：260℃；TRAP Initial temperature：40℃；TRAP final temperatures：260℃；Sample introduction needle temperature：110℃；Incubation temperature：90℃；Brooding time：1200s；During trapping Between：50s；Sample size：500μL.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610398897.7A CN105954451B (en) | 2016-06-06 | 2016-06-06 | Cigarette type Quick method based on the panchromatic modal data of electronic nose |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610398897.7A CN105954451B (en) | 2016-06-06 | 2016-06-06 | Cigarette type Quick method based on the panchromatic modal data of electronic nose |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105954451A CN105954451A (en) | 2016-09-21 |
| CN105954451B true CN105954451B (en) | 2017-12-26 |
Family
ID=56907950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610398897.7A Active CN105954451B (en) | 2016-06-06 | 2016-06-06 | Cigarette type Quick method based on the panchromatic modal data of electronic nose |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105954451B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107807199B (en) * | 2017-10-31 | 2018-08-28 | 华东理工大学 | One grows tobacco and tobacco product electronic nose instrument detection method |
| CN107843695B (en) * | 2017-10-31 | 2018-08-28 | 华东理工大学 | The electronic nose instrument evaluation method of tobacco and tobacco product aesthetic quality |
| CN109781870B (en) * | 2017-11-13 | 2022-11-22 | 贵州中烟工业有限责任公司 | Method for identifying flavor type of flue-cured tobacco |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070099239A1 (en) * | 2005-06-24 | 2007-05-03 | Raymond Tabibiazar | Methods and compositions for diagnosis and monitoring of atherosclerotic cardiovascular disease |
| CN100444153C (en) * | 2005-11-28 | 2008-12-17 | 颐中烟草(集团)有限公司 | Cigarette internal quality index extimating method based on regression function estimating SVM |
| CN104634744B (en) * | 2014-12-19 | 2017-12-01 | 云南省烟草公司昆明市公司 | A kind of Maturity of Tobacco Leaf assay method based on reflectance spectrum |
| CN104820077A (en) * | 2015-05-29 | 2015-08-05 | 广东中烟工业有限责任公司 | Method and device for recognizing cigarette |
-
2016
- 2016-06-06 CN CN201610398897.7A patent/CN105954451B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105954451A (en) | 2016-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105954451B (en) | Cigarette type Quick method based on the panchromatic modal data of electronic nose | |
| CN110057955B (en) | Method for screening specific serum marker of hepatitis B | |
| Brudzewski et al. | Metal oxide sensor arrays for detection of explosives at sub-parts-per million concentration levels by the differential electronic nose | |
| Olivier | A metabolomics approach to characterise and identify various Mycobacterium species | |
| WO2019041488A1 (en) | Toxic and harmful gas detection and recognition method based on machine olfaction | |
| Shen et al. | Determination of the geographic origin of rice by element fingerprints and correlation analyses with the soil of origin | |
| CN105512558B (en) | A kind of android ad plug-in detection methods based on decompiling modular character | |
| CN105913079A (en) | Target domain migration extreme learning-based electronic nose heterogeneous data identification method | |
| CN109002859B (en) | Sensor array feature selection and array optimization method based on principal component analysis | |
| CN110057954B (en) | Application of plasma metabolism marker in diagnosis or monitoring of HBV | |
| Rittenour et al. | Immunologic, spectrophotometric and nucleic acid based methods for the detection and quantification of airborne pollen | |
| CN109030611B (en) | Rapid and lossless green tea aroma fingerprint analysis and aroma type discrimination method based on ion mobility spectrometry | |
| CN104111274A (en) | Method for identifying producing area of red bayberry juice by using gas sensor array type electronic nose fingerprint analysis system | |
| CN107132311A (en) | A kind of fast gas recognizer extracted based on frequency domain character | |
| Choi et al. | Identification of odor emission sources in urban areas using machine learning-based classification models | |
| CN106126882A (en) | P-block element p geochemistry data method for optimizing is reconnoitred based on what Kendall's concordance coefficient sorted | |
| CN103399092B (en) | Method for rapidly evaluating quality of ginseng | |
| CN105738503B (en) | A kind of method based on electronic nose fast prediction hickory nut content of fatty acid | |
| Lee et al. | Accuracy of the high-throughput amplicon sequencing to identify species within the genus Aspergillus | |
| CN103487463A (en) | Honey detection method using grid optimization-based selection of parameters of support vector machine classifier | |
| CN107764793A (en) | Detection method of the electronic nose to aspergillus oryzae fermentation situation in bean paste yeast making process | |
| CN209727908U (en) | A kind of electronic nose equipment based on food inspection | |
| CN102282932A (en) | Method for rapidly detecting germination rate of vegetable seeds | |
| CN102798683B (en) | Universal total-component quantitative analysis method of gas chromatography-mass spectrometry | |
| CN106918817A (en) | A kind of method for optimizing of the gas sensor that can detect tea type and quality |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |