CN109324016A - A kind of determination method of redried odor type style - Google Patents
A kind of determination method of redried odor type style Download PDFInfo
- Publication number
- CN109324016A CN109324016A CN201811210097.3A CN201811210097A CN109324016A CN 109324016 A CN109324016 A CN 109324016A CN 201811210097 A CN201811210097 A CN 201811210097A CN 109324016 A CN109324016 A CN 109324016A
- Authority
- CN
- China
- Prior art keywords
- sample
- odor type
- redried
- style
- spectrum
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 235000019504 cigarettes Nutrition 0.000 claims abstract description 34
- 238000001228 spectrum Methods 0.000 claims abstract description 32
- 238000012360 testing method Methods 0.000 claims abstract description 13
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 241000208125 Nicotiana Species 0.000 claims description 46
- 235000002637 Nicotiana tabacum Nutrition 0.000 claims description 46
- 239000000796 flavoring agent Substances 0.000 claims description 20
- 239000003205 fragrance Substances 0.000 claims description 15
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 14
- 238000002329 infrared spectrum Methods 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 12
- 235000019634 flavors Nutrition 0.000 claims description 9
- 238000004519 manufacturing process Methods 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 5
- 101000596046 Homo sapiens Plastin-2 Proteins 0.000 claims description 3
- 102100035182 Plastin-2 Human genes 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 239000000203 mixture Substances 0.000 abstract description 8
- 239000002304 perfume Substances 0.000 abstract description 7
- 230000003595 spectral effect Effects 0.000 abstract description 5
- 238000009826 distribution Methods 0.000 abstract description 3
- 238000012423 maintenance Methods 0.000 abstract description 3
- 238000013139 quantization Methods 0.000 abstract description 2
- 239000000523 sample Substances 0.000 description 71
- 238000000513 principal component analysis Methods 0.000 description 8
- 238000011156 evaluation Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000010239 partial least squares discriminant analysis Methods 0.000 description 2
- 230000001953 sensory effect Effects 0.000 description 2
- 235000019505 tobacco product Nutrition 0.000 description 2
- 238000004566 IR spectroscopy Methods 0.000 description 1
- 238000012356 Product development Methods 0.000 description 1
- 238000010009 beating Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 150000002894 organic compounds Chemical class 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
Abstract
The present invention provides a kind of determination method of redried odor type style, a key point for being different from other methods is, it proposes a kind of method for selecting the typical sample with clear odor type style, can determine modeling sample on the basis of not by artificial calibration (expert smokes panel test).Based on typical sample, the spectrum that can really reflect odor type style is filtered out by feature extraction.It avoids due to the not significant caused subsequent spectral manipulation of sample itself odor type, the improper application of the methods of feature extraction；The present invention establishes the odor type model comprising domestic multiple producing region redrieds and constructs piece cigarette perfume type genre distribution figure by the quantization to piece cigarette perfume type, and to cigarette composition, the practical applications such as leaf group maintenance have certain directive significance.
Description
Technical field
The present invention relates to cigarette composition and tobacco product assessment technique fields.This method is suitable for being redried, provides
A kind of determination method of effective piece cigarette perfume type style.
Background technique
Odor type style is to characterize the important indicator of tobacco leaf and cigarette product, and show the important of industrial enterprise's Brand characteristic
Index.50-60 age, Chinese scholar propose domestic flue-cured tobacco being divided into Luzhou-flavor, flat-flavor and delicate fragrance type three categories.Later,
Someone odor type " refinement ", by three original big odor types be refined as delicate fragrance type, it is clear partially in, in partially clear, osculant, in it is partially dense, dense
Luzhou-flavor is neutralized partially.However, the judgement of odor type is affected mainly by sensory evaluating smoking by subjective factor, researchers are caused
The identification of different producing area odor type is had differences.By natural climate, the influence of planting type etc., between the different years of same producing region
Tobacco style also has fluctuation.And basic component units -- the redried of cigarette product, and pass through module formula, redrying, group
At increasingly complex compared with single-tobacco-typed cigarette, showing for odor type style is also increasingly complex.Therefore need to seek it is a kind of be detached from suction evaluation, it is objective
Decision technology, determine the odor type style of piece cigarette.
Infrared spectroscopy is widely used in industrial circle because its is quick, accurate and lossless.Spectrum multivariate calibration techniques energy
Enough it is efficiently used for material component content detection and online process monitoring.Near infrared spectrum is mainly times vibrated to hydric group
Frequency and sum of fundamental frequencies absorb, wherein the composition information of most of type organic compounds is contained, with the associated letter of tobacco leaf chemical composition
It ceases very rich, therefore redried style is predicted using near infrared spectrum, can objectively reflect each of the year
The odor type style of cigarette.Shi Fengcheng etc. using based near infrared spectrum PLS-DA algorithm differentiate the tobacco leaf place of production, to Sichuan, Yunnan,
Chongqing, 4, Fujian producing region single-tobacco-typed cigarette establish place of production discrimination model respectively, are greater than to verifying collection sample predictions precision in each producing region
93%.CN201610834599- application discloses a kind of tobacco style similitude classification method based near infrared light spectrum signature, leads to
Cross Principal Component Analysis extract spectral signature, then by rows to all near infrared spectrums, obtain data matrix, to its into
Row principal component analysis calculates the mean value of principal component by chemometrics method, determines classification line, establishes the region tobacco leaf wind
The PCA of lattice distinguishes model, and the substitution for producing region tobacco leaf in short supply provides scientific basis；And its goal of the invention be can be fast and accurately
Similitude identification is carried out to the style of tobacco leaf, this method is simple, easily operated, environment friendly and pollution-free.
[Wang Yiding, Zhao Mingqin, Fu Bo wait using the method for visible-near-infrared spectrum identification different flavor style flue-cured tobacco]
It discloses to realize the quick discrimination to flue-cured tobacco flavor types style, using visible-near-infrared spectrum technology to Luzhou-flavor, intermediate perfume
3 kinds of type, delicate fragrance type flue-cured tobacco samples carry out spectra collection, obtain the characteristic spectrum information of tobacco leaf intrinsic chemical ingredient.Choose all-wave
Section characteristic spectrum is through mean normalization, Savitzky-Golay convolution is smooth, multiplicative scatter correction and standard normal transformation etc. are excellent
Change pretreatment, clustering is carried out to spectroscopic data using principal component analysis (PCA), and combine partial least squares discriminant analysis
(PLSDA) 3 kinds of odor type style flue-cured tobacco qualitative discrimination models are established.In addition, Shi Fengcheng, Li Dongliang, Feng Guanglin etc. are based on close red
The PLS-DA algorithm of external spectrum differentiates the cured tobacco leaf place of production [J] tobacco science and technology.2013 (4) king, one fourth, Zhao Mingqin pay rich equal
Using method [J] Chinese Tobacco Science of visible-near-infrared spectrum identification different flavor style flue-cured tobacco, 2015 (6) are all referred to
Odor type style is identified by near infrared spectrum.
Above-mentioned technical proposal is analyzed mainly for the place of production, the odor type etc. of single-tobacco-typed cigarette, and odor type sentences in data
Surely it smokes panel test by expert.According to majority principle, the odor type of sample is appointed as one of faint scent, intermediate fragrant and giving off a strong fragrance.In fact,
The odor type of different tobacco leaves is not independent, and delicate fragrance type tobacco leaf may also have intermediate fragrant or giving off a strong fragrance note.For redried,
By formula and beating and double roasting, certain variation is had occurred in the chemical component of piece cigarette, and composition is increasingly complex, and a large amount of piece cigarette does not have
Have typical odor type style, even if expert opinion there is also heavy workload and it is at sixes and sevens phenomena such as.
The prior art, which relies on, to be manually specified odor type (label) and is modeled, and does not consider tobacco portion tobacco leaf odor type not allusion quotation
The actual conditions of type also lack the method for identifying these tobacco leaf samples.The use of these samples will lead to modeling inaccuracy, reach
Less than the effect of application.In addition, the analysis result of the prior art is faint scent, intermediate fragrant or giving off a strong fragrance one kind, cannot reflect not
Otherness is shown with odor type between sample, is not able to satisfy purchase evaluation, the practical businesses demand such as formula.
Summary of the invention
The present invention provides a kind of determination method of redried odor type style, and a key point for being different from other methods exists
In, propose a kind of method for selecting the typical sample with clear odor type style, it can be by artificial calibration, (expert not be commented
Inhale) on the basis of determine modeling sample.Based on typical sample, odor type style can really be reflected by being filtered out by feature extraction
Spectrum.It avoids due to the not significant caused subsequent spectral manipulation of sample itself odor type, the methods of feature extraction is no
Work as application；The present invention establishes the odor type model comprising domestic multiple producing region redrieds, passes through the quantization to piece cigarette perfume type, structure
Piece cigarette perfume type genre distribution figure is built, to cigarette composition, the practical applications such as leaf group maintenance have certain directive significance.
In order to achieve the above-mentioned object of the invention, the present invention provides a kind of determination method of redried odor type style, specific
Technical solution are as follows:
A kind of determination method of redried odor type style, includes the following steps:
Step 1) provides K redried sample, its sample spectra is acquired near infrared spectrometer, and to the light of acquisition
Spectrum carries out smooth and seeks first derivative processing；
Step 2) carries out first odor type calibration, the mode of odor type calibration are as follows: the tobacco leaf in Yunnan is labeled as to each cigarette sample
Faint scent, the place of production are that the tobacco leaf in Hunan, Henan and Anhui is labeled as giving off a strong fragrance, other tobacco leaves are labeled as intermediate fragrant；
Step 3) to piece cigarette sample carry out n times sampling, every time sampling in, in K sample in the piece cigarette of every kind of odor type with
Machine extracts the sample of same percentage, forms training set, the sample that do not draw is as test set；Every time after sampling, to training set
Model the prediction result of simultaneously labeled test collection sample；Count all samples as forecast set when prediction result, select
Typical sample；
Step 4) utilizes typical sample and its corresponding odor type, carries out the Feature Selection in spectrum；
Step 5) establishes odor type Style Model using PLS2；
Step 6) carries out secondary calibration, new modeling of laying equal stress on to modeling sample odor type；
Step 7) acquires the near infrared spectrum of sample to be tested, and the model established using step 6) is to the near-infrared of sample to be tested
Spectrum is predicted, determines the odor type style of sample to be tested.
Preferably, the extraction ratio sampled every time in step 3) is the 70%-80% of total sample number, the sample extracted every time
Sample accounting of different flavor is consistent with the odor type accounting in original sample in this.
Preferably, the modeling method in step 3) is linear discriminant analysis method, the mark mode of each test sample
Correct/error is predicted for odor type.
Preferably, statistical method in step 3) is prediction error rate p, p=prediction error number/be used as test sample
Number；Step) in typical sample select mode be p=0 sample.
Preferably, the spectrum characteristic selection mode in step 4) is the difference journey of three kinds of odor types in each spectrum wave number point
Degree, is defined as: Sb/Sw；It is different flavor inter _ class relationship matrix for same odor type within class scatter matrix:
Wherein,
X: being indicated training sample, indicated using column vector,
C: indicating C class sample,
μi: indicate the mean value (i=1,2 ..., C) of the i-th class training sample,
Sw: indicating Scatter Matrix in class,
Sb: class scatter matrix is indicated.
Preferably, the modeling method in step 5) is offset minimum binary, wherein X is the spectrum selected after characteristic point, Y
For odor type label；Wherein, delicate fragrance type is [1 0 0], and flat-flavor is [0 1 0], and Luzhou-flavor is [0 0 1]；Latent variable number by
Cross validation error and F, which are examined, to be determined, in the data, model latent variable number is 6.
Preferably, the secondary calibration of the odor type of step 6) is demarcated by the predicted value of training sample in modeling process.
Preferably, the spectrum of sample to be tested and modeling sample carry out identical pretreatment and feature selecting in step 7).
The method of the present invention has a characteristic that 1. by way of stratified sampling-analysis-secondary calibration of statistics-modeling-,
Solve the problems, such as that redried sample odor type style is indefinite；2. the sample with typical style is filtered out first, in this base
Corresponding Feature Selection and modeling are carried out on plinth, do not find this method in existing method；3. using clear three kinds of dimensions in dense
Mode characterizes the odor type style of piece cigarette, more practicability；4. the present invention provides data of smokeing panel test without expert.
Existing method only leans on sensory evaluation to the evaluation of piece cigarette perfume type style, takes time and effort when sample size is big more.
By the method for the invention, it realizes the Analysis And Evaluation to the odor type style of sheets used cigarette in formula, is product development, composition maintenance
Reference direction is provided.In addition, evaluating piece cigarette processing quality, cigarette of adjusting is evaluated, and the offers references such as tobacco leaf evaluation are expanded in new producing region
Opinion.
Detailed description of the invention
Fig. 1 is Technology Roadmap of the invention；
Fig. 2 is sheets used cigarette spectrum in embodiment；
Fig. 3 is the prediction error rate of each sample under multiple sampling；
Fig. 4 is based on the model coefficient figure after typical sample and feature extraction；
Fig. 5 is the odor type distribution map for inventory's piece cigarette that the present invention is calculated.
Specific embodiment
A kind of determination method of redried odor type style, as shown in Figure 1, specifically comprising the following steps:
Step 1) chooses certain business inventory piece cigarette 477, sample, by sample according to tobacco business standard " YC/T after sampling
The preparation and determination of moisture Oven Method of 31-1996 tobacco and tobacco product sample " it is prepared into Powder samples and (tobacco leaf is placed in baking oven
In, dry 4h at 40 DEG C, with milled 40 mesh of Cyclone mill (FOSS)), spectral measurement, training set sample are carried out after sealing and balancing 1d
This spectrogram is as shown in Figure 2；It carries out smooth to the acquisition spectrum in step 1) and seeks first derivative pretreatment, preprocess method
It is to eliminate the scattering and other interference in spectrum；
Step 2) carries out first odor type calibration, the mode of odor type calibration are as follows: the tobacco leaf in Yunnan is labeled as to each cigarette sample
Faint scent, the place of production are that the tobacco leaf in Hunan, Henan and Anhui is labeled as giving off a strong fragrance, other tobacco leaves are labeled as intermediate fragrant；To different flavor
The piece cigarette of tobacco leaf mixing composition, main odor type accounting are labeled as its main odor type, otherwise do not make marks 90% or more,
It is not involved in subsequent screening, is only predicted.
Step 3) executes 2000 sampling, randomly selects in 477 samples every time 334 (70%), when extracting every time
Guarantee that the position ratio in sample is consistent with original sample collection, LDA modeling, record are carried out to 334 samples and corresponding odor type
Remaining 143 sample predictions results；The prediction error rate (with reference to Fig. 3) for counting all samples after sampling 2000 times, is picked out
The sample that prediction error rate is 0 is as typical sample；
Step 4) carries out feature extraction to typical sample and corresponding odor type；Spectrum characteristic selection mode is each spectrum wave
The difference degree of three kinds of odor types in several points, is defined as: Sb/Sw；For same odor type within class scatter matrix, between different flavor class
Scatter matrix:
Wherein,
X: being indicated training sample, indicated using column vector,
C: indicating C class sample,
μi: indicate the mean value (i=1,2 ..., C) of the i-th class training sample,
Sw: indicating Scatter Matrix in class,
Sb: class scatter matrix is indicated.
Step 5) establishes odor type Style Model (with reference to Fig. 4) using PLS2, and modeling method is offset minimum binary, wherein X is
Spectrum after selecting characteristic point, Y are odor type label.Wherein, delicate fragrance type is [1 0 0], and flat-flavor is [0 1 0], Luzhou-flavor
For [0 0 1].Latent variable number is examined by cross validation error and F and is determined, in the data, model latent variable number is 6；
Odor type is characterized by using three-dimensional label, can more specifically symbolize the odor type difference between different samples, such as
It can symbolize based on a typical faint scent sample and a faint scent, the difference of slightly intermediate fragrant sample.In the prior art, two
A sample standard deviation is labeled as faint scent, can not show its difference.
Step 6) carries out secondary calibration, new modeling of laying equal stress on to modeling sample odor type；The secondary calibration of odor type is by modeling
The predicted value of training sample is demarcated in journey.
The effect of secondary calibration is the part expert opinion disunity removed in primitive modeling sample, the indefinite sample of odor type
This, as far as possible improves the precision of data label, to improve the availability of model.
Step 7) obtains spectrum to unknown sample in the same way, and the model established using step 6) is to sample to be tested
Near infrared spectrum predicted, determine the odor type style of sample to be tested；The spectrum of sample to be tested is identical as modeling sample carry out
Pretreatment and feature selecting.
Sample odor type is given by experience or manually given in the prior art, it is contemplated that the stability that people smokes panel test,
The subjective factors such as sensitivity and preference of smokeing panel test in the different human world, so that, always containing the interference of subjective factor, making in subsequent analysis
Reliability and the applicability for obtaining result be not strong.
The present invention selects typical sample by multiple sampling, instead of the mode manually demarcated, and people is not necessarily in modeling process
Work adjusting parameter, it is as a result more objective.By taking Fig. 5 in embodiment as an example, Shandong (SD) tobacco leaf in evaluation, do not unite by expert opinion
One, there is giving off a strong fragrance, it is intermediate fragrant and in partially dense etc. opinions.If pressing existing method, Shandong is demarcated and is built by a certain odor type
Mould, result is not inconsistent with other experts necessarily, and is difficult to explain.In the present invention, Shandong tobacco leaf is screened as atypia tobacco leaf, no
Participate in modeling.In its prediction result, Shandong tobacco leaf category Luzhou-flavor, but intensity is low compared with Hunan tobacco leaf, and in the whole country, Shandong cigarette
Leaf and Yunnan tobacco difference are maximum, and with Hunan, the giving off a strong fragrances tobacco leaf such as Henan also has certain difference.These results obtain most special
The approval of family, not only gives odor type, but also reflect difference, more meet practical application.
CN201610834599- application discloses a kind of tobacco style similitude classification side based near infrared light spectrum signature
Method extracts spectral signature by Principal Component Analysis, then by rows to all near infrared spectrums, obtains data matrix,
Principal component analysis is carried out to it, the mean value of principal component is calculated by chemometrics method, classification line is determined, establishes the region
The PCA of tobacco style distinguishes model.This method fails to screen whether model sample has typical odor type feature before modeling,
It also fails to characterize in model result and shows otherness between same odor type sample.
Claims (8)
1. a kind of determination method of redried odor type style, includes the following steps:
Step 1) provides K redried sample, acquires its sample spectra near infrared spectrometer, and to the spectrum of acquisition into
Row is smooth and seeks first derivative processing；
Step 2) carries out first odor type calibration, the mode of odor type calibration to each cigarette sample are as follows: and the tobacco leaf in Yunnan is labeled as faint scent,
The place of production is that the tobacco leaf in Hunan, Henan and Anhui is labeled as giving off a strong fragrance, other tobacco leaves are labeled as intermediate fragrant；
Step 3) carries out n times sampling to piece cigarette sample, every time in sampling, to taking out at random in the piece cigarette of every kind of odor type in K sample
The sample of same percentage is taken, forms training set, the sample that do not draw is as test set；Every time after sampling, training set is carried out
Model the prediction result of simultaneously labeled test collection sample；Count all samples as forecast set when prediction result, select typical case
Sample；
Step 4) utilizes typical sample and its corresponding odor type, carries out the Feature Selection in spectrum；
Step 5) establishes odor type Style Model using PLS2；
Step 6) carries out secondary calibration, new modeling of laying equal stress on to modeling sample odor type；
Step 7) acquires the near infrared spectrum of sample to be tested, and the model established using step 6) is to the near infrared spectrum of sample to be tested
It is predicted, determines the odor type style of sample to be tested.
2. a kind of determination method of redried odor type style as described in claim 1, it is characterised in that: in step 3) every time
The extraction ratio of sampling is the 70%-80% of total sample number, in the sample extracted every time the sample accounting of different flavor with it is original
Odor type accounting in sample is consistent.
3. a kind of determination method of redried odor type style as described in claim 1, it is characterised in that: building in step 3)
Mould method is linear discriminant analysis method, and the mark mode of each test sample is that odor type predicts correct/error.
4. a kind of determination method of redried odor type style as described in claim 1, it is characterised in that: the system in step 3)
Meter method is the number of prediction error rate p, p=prediction error number/as test sample；It is p=0 that typical sample, which selects mode,
Sample.
5. a kind of determination method of redried odor type style as described in claim 1, it is characterised in that: the light in step 4)
Spectrum signature selection mode is the difference degree of three kinds of odor types in each spectrum wave number point, is defined as: Sb/Sw；For same odor type class
Interior scatter matrix is different flavor inter _ class relationship matrix:
Wherein,
X: being indicated training sample, indicated using column vector,
C: indicating C class sample,
μi: indicate the mean value (i=1,2 ..., C) of the i-th class training sample,
Sw: indicating Scatter Matrix in class,
Sb: class scatter matrix is indicated.
6. a kind of determination method of redried odor type style as described in claim 1, it is characterised in that: building in step 5)
Mould method is offset minimum binary, wherein X is the spectrum selected after characteristic point, and Y is odor type label；Wherein, delicate fragrance type is [1
00], flat-flavor is [0 1 0], and Luzhou-flavor is [0 0 1]；Latent variable number is examined by cross validation error and F and is determined.
7. a kind of determination method of redried odor type style as described in claim 1, it is characterised in that: the odor type of step 6)
Secondary calibration is demarcated by the predicted value of training sample in modeling process.
8. a kind of determination method of redried odor type style as described in claim 1, it is characterised in that: to be measured in step 6)
The spectrum and modeling sample of sample carry out identical pretreatment and feature selecting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811210097.3A CN109324016A (en) | 2018-10-17 | 2018-10-17 | A kind of determination method of redried odor type style |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811210097.3A CN109324016A (en) | 2018-10-17 | 2018-10-17 | A kind of determination method of redried odor type style |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109324016A true CN109324016A (en) | 2019-02-12 |
Family
ID=65262562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811210097.3A Pending CN109324016A (en) | 2018-10-17 | 2018-10-17 | A kind of determination method of redried odor type style |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109324016A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110132880A (en) * | 2019-06-18 | 2019-08-16 | 浙江中烟工业有限责任公司 | A kind of tobacco leaf entirety sense organ quality evaluating method based near infrared spectrum |
| CN110132879A (en) * | 2019-06-18 | 2019-08-16 | 浙江中烟工业有限责任公司 | A kind of flue-cured tobacco note discrimination method based near infrared spectrum |
| CN110736718A (en) * | 2019-10-16 | 2020-01-31 | 浙江中烟工业有限责任公司 | Method for identifying producing area and grade of flue-cured tobacco shreds |
| CN113627647A (en) * | 2021-07-08 | 2021-11-09 | 广东省农业科学院作物研究所 | Tobacco leaf odor type prediction method based on deep learning, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040055211A1 (en) * | 2000-05-25 | 2004-03-25 | Torbjorn Lestander | Single seed sortation |
| CN103226135A (en) * | 2012-10-15 | 2013-07-31 | 红塔烟草(集团)有限责任公司 | Detection and determination method for flavor type of flue-cured tobacco |
| CN104568823A (en) * | 2015-01-07 | 2015-04-29 | 中国农业大学 | Tobacco leaf raw material proportioning ratio calculation method and tobacco leaf raw material proportioning ratio calculation device based on near infrared spectrum |
| CN106501208A (en) * | 2016-09-20 | 2017-03-15 | 广西中烟工业有限责任公司 | A kind of tobacco style similitude sorting technique based near infrared light spectrum signature |
| CN108181263A (en) * | 2017-12-29 | 2018-06-19 | 浙江中烟工业有限责任公司 | The extraction of tobacco leaf genius loci and method of discrimination based near infrared spectrum |
-
2018
- 2018-10-17 CN CN201811210097.3A patent/CN109324016A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040055211A1 (en) * | 2000-05-25 | 2004-03-25 | Torbjorn Lestander | Single seed sortation |
| CN103226135A (en) * | 2012-10-15 | 2013-07-31 | 红塔烟草(集团)有限责任公司 | Detection and determination method for flavor type of flue-cured tobacco |
| CN104568823A (en) * | 2015-01-07 | 2015-04-29 | 中国农业大学 | Tobacco leaf raw material proportioning ratio calculation method and tobacco leaf raw material proportioning ratio calculation device based on near infrared spectrum |
| CN106501208A (en) * | 2016-09-20 | 2017-03-15 | 广西中烟工业有限责任公司 | A kind of tobacco style similitude sorting technique based near infrared light spectrum signature |
| CN108181263A (en) * | 2017-12-29 | 2018-06-19 | 浙江中烟工业有限责任公司 | The extraction of tobacco leaf genius loci and method of discrimination based near infrared spectrum |
Non-Patent Citations (5)
| Title |
|---|
| 张建平 等: "烟草化学成分的近红外快速定量分析研究", 《烟草科技》 * |
| 张秋菊 等: "近红外光谱技术识别烤烟香气风格的研究", 《现代仪器与医疗》 * |
| 栾丽丽 等: "应用近红外光谱和多算法融合方法分析烤烟的香型风格特征", 《光谱学与光谱分析》 * |
| 王一丁 等: "利用可见-近红外光谱鉴定不同香型风格烤烟的方法", 《中国烟草科学》 * |
| 缪明明 等: "《不同风格特征烟叶化学成分研究》", 31 December 2016 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110132880A (en) * | 2019-06-18 | 2019-08-16 | 浙江中烟工业有限责任公司 | A kind of tobacco leaf entirety sense organ quality evaluating method based near infrared spectrum |
| CN110132879A (en) * | 2019-06-18 | 2019-08-16 | 浙江中烟工业有限责任公司 | A kind of flue-cured tobacco note discrimination method based near infrared spectrum |
| CN110132880B (en) * | 2019-06-18 | 2022-05-27 | 浙江中烟工业有限责任公司 | Tobacco leaf overall sensory quality evaluation method based on near infrared spectrum |
| CN110736718A (en) * | 2019-10-16 | 2020-01-31 | 浙江中烟工业有限责任公司 | Method for identifying producing area and grade of flue-cured tobacco shreds |
| CN113627647A (en) * | 2021-07-08 | 2021-11-09 | 广东省农业科学院作物研究所 | Tobacco leaf odor type prediction method based on deep learning, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109324016A (en) | A kind of determination method of redried odor type style | |
| CN107796782B (en) | Redrying quality stability evaluation method based on tobacco leaf characteristic spectrum consistency measurement | |
| CN108181263A (en) | The extraction of tobacco leaf genius loci and method of discrimination based near infrared spectrum | |
| CN109100321A (en) | A kind of cigarette recipe maintenance method | |
| CN109640708A (en) | One device and method for being classified as tobacco sample in one group of predetermined taste classification | |
| CN103217408B (en) | Method for identifying two flue-cured tobaccos with different odor types in Guizhou | |
| CN109975238B (en) | Substitution method of tobacco leaf and cigarette leaf group formula based on near infrared spectrum | |
| CN105445421B (en) | A kind of method of aesthetic quality during predictive slice nicotinic alcohol by appearance index | |
| Ni et al. | Pattern recognition of Chinese flue-cured tobaccos by an improved and simplified K-nearest neighbors classification algorithm on near infrared spectra | |
| CN102338780B (en) | Method for discriminating cigarette brands | |
| CN104990895B (en) | A kind of near infrared spectrum signal standards normal state bearing calibration based on regional area | |
| CN106560841A (en) | Wuyi rock tea production place identification method based on deep learning | |
| CN110307871B (en) | Method for rapidly detecting quality of traditional Chinese medicine decoction pieces by adopting electronic sensory fusion | |
| CN101419209A (en) | Cigarette sensing appraise and flume index immune neural net prediction method | |
| CN104237060A (en) | Multi-index quick detection method of honeysuckle | |
| CN106198446A (en) | The method of L-Borneol content near infrared spectrum quick test Herba Blumeae Balsamiferae leaf powder | |
| Bi et al. | Quality evaluation of flue-cured tobacco by near infrared spectroscopy and spectral similarity method | |
| CN105138834A (en) | Tobacco chemical value quantifying method based on near-infrared spectrum wave number K-means clustering | |
| CN110749565A (en) | Method for rapidly identifying storage years of Pu' er tea | |
| CN106560699A (en) | Combined detection method used for identification of producing area of Wuyi rock tea | |
| CN110132880A (en) | A kind of tobacco leaf entirety sense organ quality evaluating method based near infrared spectrum | |
| CN110132879A (en) | A kind of flue-cured tobacco note discrimination method based near infrared spectrum | |
| CN106770607B (en) | A method of identifying genuine-fake cigarette using HS-IMR-MS | |
| Xiang et al. | Simultaneous identification of geographical origin and grade of flue-cured tobacco using NIR spectroscopy | |
| CN104596980A (en) | Method for measuring hot water solvends of reconstituted tobacco by paper-making process by virtue of near infrared reflectance spectroscopy technique |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190212 |