CN104899424A - Microscopic feature based method for objectively testing ripeness and leaf structure of flue-cured tobacco leaves - Google Patents
Microscopic feature based method for objectively testing ripeness and leaf structure of flue-cured tobacco leaves Download PDFInfo
- Publication number
- CN104899424A CN104899424A CN201510227248.6A CN201510227248A CN104899424A CN 104899424 A CN104899424 A CN 104899424A CN 201510227248 A CN201510227248 A CN 201510227248A CN 104899424 A CN104899424 A CN 104899424A
- Authority
- CN
- China
- Prior art keywords
- tobacco leaf
- model
- cell
- ripeness
- blade construction
- 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.)
- Granted
Links
Landscapes
- Manufacture Of Tobacco Products (AREA)
- Manufacturing Of Cigar And Cigarette Tobacco (AREA)
Abstract
A microscopic feature based method for objectively testing the ripeness and the leaf structure of flue-cured tobacco leaves, characterized by comprising: performing expert quantitative evaluation on the ripeness and the leaf structure of modeled tobacco leaves, and using instruments to measure the three microscopic feature parameters: the area of the cell, the perimeter of the cell, and the shape factor of the cell; using a stepwise regression method to establish a microscopic feature parameter-based inversion model of an expert evaluation result of the ripeness and the leaf structure of flue-cured tobacco leaves; and measuring the microscopic feature parameters of to-be-tested tobacco leaves, inputting measurements into the model, and implementing objective evaluation on the ripeness and the leaf structure of the tobacco leaves. The main advantages of the method of the present invention lie in that: 1) determining is performed objectively, wherein microscopic feature parameters are measured by using instruments and are input into the model to obtain the evaluation result of the ripeness and the leaf structure of tobacco leaves, thereby preventing subjective factors from acting; 2) evaluation of the ripeness and the leaf structure of tobacco leaves can be performed without the need for professional technicians that are qualified to evaluate the appearance quality of tobacco leaves; and 3) simplicity and convenience are achieved, the prediction error of the model is small, and the prediction accuracy of the model is desirable.
Description
Technical field
The present invention relates to tobacco leaf Evaluation on Appearance Quality field, be specially a kind of cured tobacco leaf degree of ripeness based on microscopic feature and blade construction method for objectively evaluating.
Background technology
Tobacco leaf presentation quality is one of main contents of quality of tobacco evaluation, is also be connected China's leaf tobacco production (tobacco grower, purchasing station) allots (Cigarette Industrial Enterprise) link principal element to industry at present.Degree of ripeness and blade construction are the importances of tobacco leaf presentation quality, it is generally acknowledged, the tobacco leaf judging parameter that degree of ripeness is high, blade construction is loose is better.Degree of ripeness, blade construction are main relevant with cell development situation, and the tobacco leaf cells that degree of ripeness is good reaches the state of extending, and blade construction is then closely related with the density degree of cell arrangement.At present for the evaluation Main Basis flue-cured tobacco grade scale GB2635-92 of tobacco leaf quality index appearance, evaluated by flue-cured tobacco classification expert, comprise qualitative evaluation (each index different class tobacco leaf proportion) and quantitatively evaluating (each index carries out the quantification tax point of 0 ~ 10 point according to quality), the requirement of this evaluation method to the personnel of evaluation is higher, abundant tobacco leaf Evaluation on Appearance Quality experience and relevant qualification need be possessed, and evaluation result is subject to subjective factor impact, comparability and the reappearance of evaluation result are general.Therefore, the method for objectively evaluating exploring tobacco leaf quality index appearance is the hot issue of this area always.At present the color in tobacco leaf appearance index, oil content are had to the research of part method for objectively evaluating, tobacco leaf color is judged as utilized color model (RGB, Lab), near-infrared model is evaluated tobacco leaf oil and is graded, and there is not been reported for the method for objectively evaluating of degree of ripeness and blade construction.
Summary of the invention
The object of the present invention is to provide a kind of degree of ripeness and blade construction method for objectively evaluating of cured tobacco leaf, address the deficiencies of the prior art.
Object of the present invention is achieved through the following technical solutions:
In laboratory, modeling tobacco leaf is carried out to expert's quantitatively evaluating of degree of ripeness and blade construction, adopt the cell perimeter of scanning electron microscope and image processing software measurement modeling tobacco leaf, cell area and the cell shape factor afterwards, obtain tobacco leaf microscopic feature parameter.Respectively with degree of ripeness, blade construction or degree of ripeness+blade construction expert opinion score value for dependent variable, any indicator combination in cell perimeter, cell area, the single index of the cell shape factor or this 3 indexs is that independent variable carries out stepwise regression analysis modeling, by the coefficient of determination, the significance test result of more different regression model, with cell area, cell perimeter and the cell shape factor for independent variable, can realize preferably adding Maturity of Tobacco Leaf and blade construction score value and regression modeling.Utilize this model, do not need the personnel possessing tobacco leaf Evaluation on Appearance Quality specialized capability, the evaluation to Maturity of Tobacco Leaf and blade construction can be realized.
Concrete steps are as follows:
(1) expert opinion of modeling tobacco leaf sample presentation quality.By tobacco leaf to be measured equilibrium moisture content to 15% ~ 16% under constant temperature and humidity condition, in laboratory conditions, by flue-cured tobacco classification expert, expert's quantitatively evaluating is carried out to the degree of ripeness of modeling tobacco leaf and blade construction, the Main Basis of presentation quality expert opinion is flue-cured tobacco grade scale GB2635-92, carries out evaluation give a mark by flue-cured tobacco classification expert (the senior technician of the classification assert by tobacco business or national classification contest technical expert).
(2) modeling tobacco leaf microscopic feature parameter measurement.Utilize scanning electron microscope to obtain the surface microscopic characteristic image of modeling tobacco leaf, utilize image processing software to measure cell perimeter, cell area and the cell shape factor, obtain the microscopic feature parameter of modeling tobacco leaf; The cell shape factor is calculated by [cell perimeter 2 ÷ (4 × π × cell area × 1.064)].
(3) model construction.Respectively with degree of ripeness, blade construction or degree of ripeness+blade construction expert opinion score value for dependent variable, any indicator combination in cell perimeter, cell area, the single index of the cell shape factor or this 3 indexs is that independent variable carries out stepwise regression analysis modeling, by the coefficient of determination, the significance test result of more different regression model, with cell area, cell perimeter and the cell shape factor for independent variable, can realize preferably adding Maturity of Tobacco Leaf and blade construction score value and regression modeling, build model as follows:
Y (degree of ripeness+blade construction score value)=
16.814-0.006 × cell areas (μm 2)+0.123 × cell perimeter (μm)-8.318 × cell shape factor
(4) model testing.Utilize respectively and participate in modeling sample and the outer tobacco leaf sample of modeling, the tobacco leaf microstructural parameter of apparatus measures is brought into the model that step (3) builds, output degree of ripeness and blade construction predict the outcome.According to Maturity of Tobacco Leaf and blade construction expert opinion result, model Output rusults is tested, as calculated, the model output valve of modeling sample and the average relative error of expert opinion value are 3.18%, and the outer model output valve of sample of model and the average relative error of expert opinion value are 6.20%.
(5) models applying: measure the cell perimeter of tobacco leaf to be measured, cell area and the cell shape factor according to step 2, and result is brought into step 3 established model, the evaluation result of Maturity of Tobacco Leaf and blade construction can be drawn.
The main advantage of the inventive method is: 1. objective judgement, utilizes the microscopic feature parameters input model of apparatus measures, obtains Maturity of Tobacco Leaf and blade construction evaluation result, avoids subjective factor to affect; 2. do not need the professional and technical personnel with tobacco leaf Evaluation on Appearance Quality quality can realize the evaluation of Maturity of Tobacco Leaf and blade construction; 3, simple and convenient, the prediction deviation of model is lower, and the accuracy of model prediction is better.
Embodiment
The present invention does to analyze further below in conjunction with embodiment:
Within (1) 2014 year, gather 21,3 producing region cured tobacco leaf samples such as Baoshan, Yunnan, Zhaotong and Xuchang, Henan.
(2) to be composed point according to the quantification that the quality of each sample degree of ripeness and blade construction carries out 0 ~ 10 point by flue-cured tobacco classification expert.
(3) adopt scanning electron microscope to obtain the surface microstructure image of each sample, adopt PHOTOSHOP software processing method, measure cell area, cell perimeter and the cell shape factor.
(4) use correlation analysis and stepwise regression analysis, analyze microscopic feature parameter and Maturity of Tobacco Leaf and blade construction expert and quantize to compose the quantitative relationship of dividing, and build optimum regression forecast model:
Y (degree of ripeness+blade construction score value)=
16.814-0.006 × cell areas (μm 2)+0.123 × cell perimeter (μm)-8.318 × cell shape factor
Regression model assay is in table 1.
the Parameters in Regression Model assay of table 1 Maturity of Tobacco Leaf and blade construction expert opinion result
(5) the microscopic feature parameter of the outer tobacco leaf sample of modeling tobacco leaf sample and model is utilized to bring regression model in step 4 into respectively, export Maturity of Tobacco Leaf and blade construction score value, and compare mean absolute error and the relative error of model output valve and expert opinion value, the results are shown in Table 2.Because expert is when carrying out the quantitatively evaluating of Maturity of Tobacco Leaf and blade construction, the unit of scoring of single index is 0.5, and therefore the prediction deviation of model is at 0.5 ~ 1.0 score unit, and the accuracy of model prediction is better.
(6) by the cell perimeter of any tobacco leaf of apparatus measures, cell area and the cell shape factor, Maturity of Tobacco Leaf and blade construction evaluation of estimate is exported by this method.
application example:
Gather Chenzhou, Hunan Province C3F grade tobacco sample, after measured, cell perimeter is 393.30 μm, and cell area is 4198.31 μm 2, and the cell shape factor is 2.76 as calculated, substitutes into model formation:
Y (degree of ripeness+blade construction score value)=
16.814-0.006 × cell areas (μm 2)+0.123 × cell perimeter (μm)-8.318 × cell shape factor
Finally show that Y=17.0 divides.
Claims (5)
1. the cured tobacco leaf degree of ripeness based on microscopic feature and blade construction objective measure, it is characterized in that: by flue-cured tobacco classification expert, expert's quantitatively evaluating is carried out to modeling Maturity of Tobacco Leaf and blade construction in laboratory, measure the cell perimeter of modeling tobacco leaf, cell area and the cell shape factor afterwards, obtain tobacco leaf microscopic feature parameter, use stepwise regression analysis method, set up the model utilizing microscopic feature parametric inversion Maturity of Tobacco Leaf and blade construction expert opinion result; Measure the microscopic feature parameter of any tobacco leaf, input model, can realize the objective evaluation to Maturity of Tobacco Leaf and blade construction integrated status, concrete steps are as follows:
(1) modeling tobacco leaf presentation quality expert opinion: choose cured tobacco leaf sample, carries out quantitatively evaluating by flue-cured tobacco classification expert to Maturity of Tobacco Leaf and blade construction;
(2) modeling tobacco leaf microscopic feature parameter measurement: integrating step (1), after carrying out presentation quality expert opinion, utilize scanning electron microscope and image processing software to measure tobacco leaf cells girth, cell area and the cell shape factor, obtain the microscopic feature parameter of modeling tobacco leaf;
(3) model construction: adopt stepwise regression method, builds the model of tobacco leaf microscopic feature parametric inversion degree of ripeness and blade construction expert quantitative evaluation result:
Y (degree of ripeness+blade construction score value)=
16.814-0.006 × cell areas (μm 2)+0.123 × cell perimeter (μm)-8.318 × cell shape factor
(4) model testing: utilize modeling sample and the outer sample of model respectively, the tobacco leaf microscopic feature parameter of measurement is brought into the model that step (3) builds, export the model evaluation result of degree of ripeness and blade construction; According to Maturity of Tobacco Leaf and blade construction expert opinion result, model evaluation result is tested, as calculated, the model output valve of modeling tobacco leaf sample and the average relative error of expert opinion value are 3.18%, and the outer model output valve of tobacco leaf sample of model and the average relative error of expert opinion value are 6.20%;
(5) models applying: measure the cell perimeter of tobacco leaf to be measured, cell area and the cell shape factor according to step 2, and result is brought into step 3 established model, degree of ripeness and the blade construction evaluation result of any tobacco leaf can be drawn.
2. detection method according to claim 1, is characterized in that: the cell shape factor in step 2 is calculated by [cell perimeter 2 ÷ (4 × π × cell area × 1.064)].
3. detection method according to claim 1, it is characterized in that: the Main Basis of the presentation quality expert opinion in step (1) is flue-cured tobacco grade scale GB2635-92, presentation quality expert is the senior technician of classification or national classification contest technical expert that are assert by tobacco business.
4. detection method according to claim 1, it is characterized in that: in model construction, respectively with degree of ripeness, blade construction or degree of ripeness+blade construction expert opinion score value are dependent variable, cell perimeter, cell area, any indicator combination in the single index of the cell shape factor or this 3 indexs is that independent variable carries out stepwise regression analysis modeling, by the coefficient of determination of more different regression model, significance test result, with cell area, cell perimeter and the cell shape factor are independent variable, and then realize adding Maturity of Tobacco Leaf and blade construction score value and regression modeling.
5. detection method according to claim 1, is characterized in that: image processing software used is Photoshop software.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510227248.6A CN104899424B (en) | 2015-05-07 | 2015-05-07 | A kind of cured tobacco leaf maturity and blade construction objective measure based on microscopic feature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510227248.6A CN104899424B (en) | 2015-05-07 | 2015-05-07 | A kind of cured tobacco leaf maturity and blade construction objective measure based on microscopic feature |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104899424A true CN104899424A (en) | 2015-09-09 |
| CN104899424B CN104899424B (en) | 2018-01-02 |
Family
ID=54032086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510227248.6A Active CN104899424B (en) | 2015-05-07 | 2015-05-07 | A kind of cured tobacco leaf maturity and blade construction objective measure based on microscopic feature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104899424B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109540894A (en) * | 2018-12-17 | 2019-03-29 | 云南省烟草公司红河州公司 | A kind of lossless rapid detection method of cured tobacco leaf maturity |
| CN112067560A (en) * | 2020-08-06 | 2020-12-11 | 红云红河烟草(集团)有限责任公司 | Method for measuring stability of tobacco feed liquid based on combination of colorimetric value and entropy weight method |
| CN109738438B (en) * | 2018-12-29 | 2021-07-23 | 扬州大学 | Method for rapidly measuring aging degree of wheat leaves |
| CN113945563A (en) * | 2021-10-19 | 2022-01-18 | 中国烟草总公司湖北省公司 | Method for evaluating quality uniformity of cigar wrapper tobacco leaves |
| CN113989176A (en) * | 2020-07-10 | 2022-01-28 | 福建省烟草公司龙岩市公司 | Construction method of fresh tobacco leaf maturity determination model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102607475A (en) * | 2012-04-10 | 2012-07-25 | 中国烟草总公司郑州烟草研究院 | Method for measuring baked tobacco cell area by using Photoshop software |
| CN102645444A (en) * | 2012-04-10 | 2012-08-22 | 中国烟草总公司郑州烟草研究院 | Method for quantificationally characterizing cell shape features of flue-cured tobacco leaves by utilizing Photoshop software |
-
2015
- 2015-05-07 CN CN201510227248.6A patent/CN104899424B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102607475A (en) * | 2012-04-10 | 2012-07-25 | 中国烟草总公司郑州烟草研究院 | Method for measuring baked tobacco cell area by using Photoshop software |
| CN102645444A (en) * | 2012-04-10 | 2012-08-22 | 中国烟草总公司郑州烟草研究院 | Method for quantificationally characterizing cell shape features of flue-cured tobacco leaves by utilizing Photoshop software |
Non-Patent Citations (2)
| Title |
|---|
| 过伟民等: "典型烤烟产区初烤烟叶表面细胞形态特征分析", 《烟草农学》 * |
| 黄清芬等: "烟叶评吸质量与外观质量的相关性分析", 《江西农业学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109540894A (en) * | 2018-12-17 | 2019-03-29 | 云南省烟草公司红河州公司 | A kind of lossless rapid detection method of cured tobacco leaf maturity |
| CN109738438B (en) * | 2018-12-29 | 2021-07-23 | 扬州大学 | Method for rapidly measuring aging degree of wheat leaves |
| CN113989176A (en) * | 2020-07-10 | 2022-01-28 | 福建省烟草公司龙岩市公司 | Construction method of fresh tobacco leaf maturity determination model |
| CN113989176B (en) * | 2020-07-10 | 2024-05-31 | 福建省烟草公司龙岩市公司 | Construction method of tobacco freshness She Chengshou degree judgment model |
| CN112067560A (en) * | 2020-08-06 | 2020-12-11 | 红云红河烟草(集团)有限责任公司 | Method for measuring stability of tobacco feed liquid based on combination of colorimetric value and entropy weight method |
| CN112067560B (en) * | 2020-08-06 | 2023-11-03 | 红云红河烟草(集团)有限责任公司 | Tobacco material liquid stability determination method based on chromaticity value and entropy weight method |
| CN113945563A (en) * | 2021-10-19 | 2022-01-18 | 中国烟草总公司湖北省公司 | Method for evaluating quality uniformity of cigar wrapper tobacco leaves |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104899424B (en) | 2018-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104899424A (en) | Microscopic feature based method for objectively testing ripeness and leaf structure of flue-cured tobacco leaves | |
| CN105445421B (en) | A kind of method of aesthetic quality during predictive slice nicotinic alcohol by appearance index | |
| CN106680137B (en) | A method of evaluation cigarette measurement of water ratio uncertainty | |
| CN111721715A (en) | Method for measuring tobacco shred blending uniformity based on combination of colorimetric value and entropy weight method | |
| CN109211829A (en) | A method of moisture content in the near infrared spectroscopy measurement rice based on SiPLS | |
| CN105843870B (en) | The analysis method and its application of repeatability and reproducibility | |
| CN103344572A (en) | Method for evaluating blending homogeneity of cut rolled stems and regenerated cut tobaccos in cigarettes | |
| CN102749140A (en) | Tobacco leaf surface color uniformity judging method of flue-cured tobacco | |
| CN105203498A (en) | Near infrared spectrum variable selection method based on LASSO | |
| CN101387624B (en) | Diagnostic method for essential chemical component of flue-cured tobacco | |
| CN104155264A (en) | Method for predicting content of turpentine in loblolly pine gum by using near infrared spectroscopy | |
| CN110632068A (en) | Method for measuring processing resistance of tobacco leaves | |
| CN105699239A (en) | Method for analyzing moisture retention ability of tobaccos and tobacco products by aid of near-infrared spectral models | |
| CN105628646B (en) | A kind of online tar prediction of cigarette and method for early warning | |
| CN101685092A (en) | Method for judging alcoholization quality of flue-cured tobacco by pH detection value of tobacco leaf | |
| CN109540837A (en) | The method that near-infrared quickly detects Boehmeria nivea leaves wood fibre cellulose content | |
| CN104596976A (en) | Method for determining protein of paper-making reconstituted tobacco through ear infrared reflectance spectroscopy technique | |
| CN103048290A (en) | Method for detecting filling value of cut tobacco | |
| CN1619306A (en) | Rice amylose content and dextrinization temperature fast synergistic determining method | |
| CN105158042A (en) | Measurement and representation method for softness of tobacco leaves | |
| CN104568828A (en) | Method for determining tensile strength of reproduced tobacco leaves of papermaking method by near-infrared diffuse reflection spectrum | |
| CN107238557A (en) | A kind of method of utilization near infrared spectroscopy quick detection calcium carbonate particle diameter distribution | |
| CN106092960A (en) | A kind of quick correction near infrared gear also detects the method for chemical composition in agricultural product | |
| CN110333177B (en) | Cigarette segmented suction resistance nondestructive rapid detection evaluation method based on cigarette paper air permeability stability | |
| CN112485372A (en) | Method for evaluating miscellaneous gas in flue gas |
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 | ||
| GR01 | Patent grant |