Methods for identifying tobacco origin

Abstract

The present disclosure relates to a tobacco origin identification method, which comprises the following steps: S10: using a Raman spectrometer to detect a tobacco sample to be tested, and obtaining a Raman spectrum curve about Raman shift-Raman intensity; S20: selecting a Raman spectrum curve segment that can characterize the tobacco origin on the Raman spectrum curve; S30: connecting two end points of the Raman spectrum curve segment to obtain a straight line segment; S40: obtaining a Raman parameter number set I of the tobacco sample to be tested according to the difference information of the Raman intensity between a point set on the Raman spectrum curve segment and a corresponding point set on the straight line segment. op ; S50: The Raman parameter set I of the tobacco sample to be tested op The Raman parameter sets I′ of tobacco samples with known origins are op Compare one by one, and select the origin of the tobacco sample with the closest Raman parameter set to the tobacco sample to be tested as the origin of the tobacco sample to be tested. By analyzing the tobacco Raman spectrum parameters, the tobacco origin can be quickly classified with high accuracy.

Description

Technical Field

[0001] The present disclosure relates to the technical field of tobacco detection, and in particular to a method for identifying the origin of tobacco. Background Art

[0002] Tobacco is an important cash crop in my country. The quality of tobacco has a great impact on the sales of cigarettes. The quality of tobacco leaves is closely related to the environment of the origin. There are obvious differences between tobacco in different regions of my country. Accurately distinguishing the origin of tobacco plays an important role in evaluating the quality of tobacco and its subsequent industrial application.

[0003] At present, the identification of tobacco origin mainly relies on subjective judgment based on expert experience and summarized appearance regional features, or the use of near-infrared spectroscopy technology combined with least squares support vector machine to identify the origin of flue-cured tobacco leaves. Judgment based on experience is highly subjective and prone to inaccurate test results. Near-infrared spectroscopy has the problem of harsh spectral detection conditions, making it difficult to achieve rapid classification of tobacco origin. Summary of the invention

[0004] Some embodiments of the present disclosure provide a tobacco origin identification method for alleviating the problem of inaccurate tobacco origin identification.

[0005] In one aspect of the present disclosure, a method for identifying the origin of tobacco is provided, comprising the following steps:

[0006] S10: testing the tobacco sample to be tested by using a Raman spectrometer to obtain a Raman spectrum curve of Raman shift-Raman intensity;

[0007] S20: selecting a Raman spectrum curve segment that can characterize the origin of tobacco from the Raman spectrum curve;

[0008] S30: connecting two end points of the Raman spectrum curve segment to obtain a straight line segment;

[0009] S40: Obtaining the Raman parameter set I of the tobacco sample to be tested according to the difference information of the Raman intensity between the point set on the Raman spectrum curve segment and the corresponding point set on the straight line segment op ;

[0010] S50: The Raman parameter set I of the tobacco sample to be tested op The Raman parameter sets I′ of tobacco samples with known origins are op Compare one by one, and select the origin corresponding to the tobacco sample of known origin that is closest to the Raman parameter number set of the tobacco sample to be tested as the origin of the tobacco sample to be tested.

[0011] In some embodiments, in step S20, the Raman shift corresponding to the Raman spectrum curve segment selected to characterize the origin of tobacco is 2700 cm-1 ~3500cm -1 .

[0012] In some embodiments, step S40 includes:

[0013] Select multiple Raman shifts on the Raman spectrum curve segment, and subtract the Raman intensity corresponding to the Raman shift on the straight line segment from the Raman intensity corresponding to the multiple Raman shifts to obtain the Raman peak number set I R ;

[0014] Select the maximum peak I on the Raman spectrum curve segment max ;

[0015] The Raman peak number set I R Each Raman peak in is divided by I max , obtain the Raman parameter set I of the tobacco sample to be tested op .

[0016] In some embodiments, in step S50, the Raman parameter set I of the tobacco sample to be tested is op The Raman parameter sets I′ of tobacco samples with known origins are op Compare them one by one, including:

[0017] According to the formula: Calculate to obtain a×b D values, and sort the a×b D values ​​in order of size;

[0018] Where a represents the number of known origins, b represents the number of tobacco samples in each known origin, and n represents the Raman parameter set I of the tobacco samples to be tested. op The number of Raman parameters in I opi Represents the Raman parameter set I of the tobacco sample to be tested op The ith Raman parameter in opi The Raman parameter set I′ represents the tobacco sample with known origin op The i-th Raman parameter in .

[0019] In some embodiments, in step S50, selecting the origin corresponding to the tobacco sample of known origin that is closest to the Raman parameter number set of the sample to be tested as the origin of the tobacco sample to be tested includes:

[0020] The smallest D value is selected from a×b D values, and the tobacco sample of known origin corresponding to the smallest D value is determined. The origin corresponding to the tobacco sample of known origin is the origin of the tobacco sample to be tested.

[0021] In some embodiments, m smallest D values ​​are selected from a×b D values, where m>a and m is an odd number. The known origin corresponding to the m smallest D values ​​has the largest number of tobacco samples, and this known origin is the origin of the tobacco samples to be tested.

[0022] In some embodiments, the value of b is greater than or equal to five.

[0023] In some embodiments, in step S50, the Raman parameter set I′ of at least one tobacco sample among a plurality of tobacco samples of known origins is op The methods for obtaining include:

[0024] S100: testing the tobacco sample using a Raman spectrometer to obtain a Raman spectrum curve of Raman shift-Raman intensity;

[0025] S200: selecting a Raman spectrum curve segment that can characterize the origin of tobacco from the Raman spectrum curve;

[0026] S300: connecting two end points of the Raman spectrum curve segment to obtain a straight line segment;

[0027] S400: Obtaining the Raman parameter set I′ of the tobacco sample according to the difference information of the Raman intensity between the point set on the Raman spectrum curve segment and the corresponding point set on the straight line segment op .

[0028] In some embodiments, in step S10, the Raman spectrometer used includes a Fourier transform Raman spectrometer; and / or the wavelength of the laser of the Raman spectrometer is selected to be 785nm or 1064nm; and / or the laser power of the laser of the Raman spectrometer is selected to be 10mW to 150mW; and / or the test spectrum superposition times of the Raman spectrometer are set to 300 times or more.

[0029] In some embodiments, step S10 includes: selecting more than eight position points of the tobacco sample to be tested, using a Raman spectrometer to detect the more than eight position points, obtaining more than eight groups of values, and obtaining a Raman spectrum curve about Raman shift-Raman intensity based on the average value of the more than eight groups of values.

[0030] Based on the above technical solution, the present disclosure has at least the following beneficial effects:

[0031] The tobacco origin identification method provided by the embodiment of the present disclosure realizes rapid classification and identification of tobacco origin with high accuracy by analyzing tobacco Raman spectrum parameters. Moreover, by obtaining the difference information of Raman intensity between a set of points on a Raman spectrum curve segment and a corresponding set of points on a straight line segment, a set of Raman parameter numbers of the tobacco sample to be tested is obtained according to the difference information, thereby effectively removing the interference of the fluorescence signal on the properties of the tobacco origin, and realizing rapid and accurate classification of the tobacco origin. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present application. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:

[0033] Figure 1 A schematic diagram of a process for identifying the origin of tobacco provided according to some embodiments of the present disclosure;

[0034] Figure 2 A schematic diagram of a Raman spectrum curve segment regarding Raman shift-Raman intensity provided according to some embodiments of the present disclosure.

[0035] It should be understood that the size of each part shown in the accompanying drawings is not drawn according to the actual proportional relationship. In addition, the same or similar reference numerals represent the same or similar components. DETAILED DESCRIPTION

[0036] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The description of the exemplary embodiments is merely illustrative and is in no way intended to limit the present disclosure and its application or use. The present disclosure can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to make the present disclosure thorough and complete and to fully express the scope of the present disclosure to those skilled in the art. It should be noted that unless otherwise specifically stated, the relative arrangement of the components and steps, the composition of the materials, the numerical expressions and the numerical values ​​set forth in these embodiments should be interpreted as being merely exemplary and not as limiting.

[0037] The words "first", "second" and similar words used in this disclosure do not indicate any order, quantity or importance, but are only used to distinguish different parts. The words "include" or "comprise" and similar words mean that the elements before the word include the elements listed after the word, and do not exclude the possibility of including other elements. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0038] In the present disclosure, when a specific device is described as being located between a first device and a second device, there may or may not be an intermediate device between the specific device and the first device or the second device. When a specific device is described as being connected to other devices, the specific device may be directly connected to the other device without an intermediate device, or may not be directly connected to the other device but have an intermediate device.

[0039] All terms (including technical terms or scientific terms) used in the present disclosure have the same meanings as those understood by ordinary technicians in the field to which the present disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in general dictionaries, such as general dictionaries, should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined herein.

[0040] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0041] Figure 1 1 is a flow chart of some embodiments of the tobacco origin identification method disclosed in the present invention. Figure 1 In some embodiments, the tobacco origin identification method comprises the following steps:

[0042] S10: testing the tobacco sample to be tested by using a Raman spectrometer to obtain a Raman spectrum curve of Raman shift-Raman intensity;

[0043] S20: selecting a Raman spectrum curve segment that can characterize the origin of tobacco from the Raman spectrum curve;

[0044] S30: Connect the two end points of the Raman spectrum curve segment to obtain a straight line segment AB, reference Figure 2 ;

[0045] S40: Obtaining the Raman parameter set I of the tobacco sample to be tested according to the difference information of the Raman intensity between the point set on the Raman spectrum curve segment and the corresponding point set on the straight line segment AB op ;

[0046] S50: The Raman parameter set I of the tobacco sample to be tested op The Raman parameter sets I′ of tobacco samples with known origins are op Compare one by one, and select the origin corresponding to the tobacco sample of known origin that is closest to the Raman parameter number set of the tobacco sample to be tested as the origin of the tobacco sample to be tested.

[0047] The tobacco origin identification method provided in the embodiment of the present disclosure analyzes tobacco Raman spectral parameters to achieve rapid classification of tobacco origin with high accuracy, without the need for complex manual classification processes, and solves the time-consuming problem of origin classification in actual industrial production analysis.

[0048] Since tobacco has fluorescence characteristics, which are the chemical properties of tobacco itself, the general trend of the Raman spectrum curve is to go diagonally downward along the increase of Raman shift. The fluorescence signal generated by the fluorescence characteristics of tobacco will cover the characteristics of the tobacco itself, affecting the Raman spectrum curve of the tobacco, and further affecting the attribute analysis of the tobacco production area. Therefore, the embodiment of the present disclosure connects the two end points of the Raman spectrum curve segment to obtain a straight line segment, which is basically a line segment generated by the fluorescence signal, and then obtains the difference information of the Raman intensity between the point set on the Raman spectrum curve segment and the corresponding point set on the straight line segment, and obtains the Raman parameter number set I of the tobacco sample to be tested according to the difference information. op , which can effectively remove the interference of fluorescence signals on the properties of tobacco origin and achieve accurate classification of tobacco origin.

[0049] In some embodiments, in step S20, the Raman shift corresponding to the Raman spectrum curve segment selected to characterize the origin of tobacco is 2700 cm -1 ~3500cm -1 The Raman spectrum curve is a second-order Raman spectrum curve about Raman shift and Raman intensity.

[0050] Raman shift is 2700cm -1 ~3500cm -1 The tobacco Raman spectrum contains a large amount of chemical structure information that determines the origin of tobacco, and is highly correlated with the origin of tobacco. Selecting this range can achieve better identification results.

[0051] In some embodiments, step S40 includes:

[0052] Select multiple Raman shifts on the Raman spectrum curve segment, and subtract the Raman intensity corresponding to the Raman shift on the straight line segment from the Raman intensity corresponding to the multiple Raman shifts to obtain the Raman peak number set I R ;

[0053] Select the maximum peak I on the Raman spectrum curve segment max ;

[0054] The Raman peak number set I R Each Raman peak in is divided by I max , obtain the Raman parameter set I of the tobacco sample to be tested op .

[0055] The multiple Raman shifts selected on the Raman spectrum curve segment may be tens, hundreds, one thousand or several thousand, or even more, so as to accurately classify the origin of tobacco.

[0056] In some embodiments, in step S50, the Raman parameter set I of the tobacco sample to be tested is op The Raman parameter sets I′ of tobacco samples with known origins are op Compare them one by one, including:

[0057] According to the formula: Calculate to obtain a×b D values, and sort the a×b D values ​​in order of size;

[0058] Wherein, a represents the number of known origins, b represents the number of tobacco samples in each known origin (the number of tobacco samples in each known origin may be the same or different), and n represents the Raman parameter number set I of the tobacco sample to be tested. op The number of Raman parameters in (that is, the number of Raman shifts selected on the Raman spectrum curve segment), I opi Represents the Raman parameter set I of the tobacco sample to be tested op The ith Raman parameter in opi The Raman parameter set I′ represents the tobacco sample with known origin op The i-th Raman parameter in .

[0059] In some embodiments, in step S50, selecting the origin corresponding to the tobacco sample of known origin that is closest to the Raman parameter number set of the sample to be tested as the origin of the tobacco sample to be tested includes:

[0060] The smallest D value is selected from a×b D values, and the tobacco sample of known origin corresponding to the smallest D value is determined. The origin corresponding to the tobacco sample of known origin is the origin of the tobacco sample to be tested.

[0061] In some embodiments, m smallest D values ​​are selected from a×b D values, where m>a and m is an odd number. The known origin corresponding to the m smallest D values ​​has the largest number of tobacco samples, and this known origin is the origin of the tobacco samples to be tested.

[0062] In some embodiments, the value of b is greater than or equal to five.

[0063] In some embodiments, in step S50, the Raman parameter set I′ of at least one tobacco sample among a plurality of tobacco samples of known origins op The methods for obtaining include:

[0064] S100: testing the tobacco sample using a Raman spectrometer to obtain a Raman spectrum curve of Raman shift-Raman intensity;

[0065] S200: selecting a Raman spectrum curve segment that can characterize the origin of tobacco from the Raman spectrum curve;

[0066] S300: connecting two end points of the Raman spectrum curve segment to obtain a straight line segment;

[0067] S400: Obtaining the Raman parameter set I′ of the tobacco sample according to the difference information of the Raman intensity between the point set on the Raman spectrum curve segment and the corresponding point set on the straight line segment op .

[0068] In some embodiments, in step S200, the Raman shift corresponding to the Raman spectrum curve segment selected to characterize the origin of tobacco is 2700 cm -1 ~3500cm -1 .

[0069] In some embodiments, step S400 includes:

[0070] Select multiple Raman shifts on the Raman spectrum curve segment, and subtract the Raman intensity corresponding to the Raman shift on the straight line segment from the Raman intensity corresponding to the multiple Raman shifts to obtain the Raman peak number set I′ R ;

[0071] Select the maximum peak I′ on the Raman spectrum curve segment max ;

[0072] The Raman peak number set I′ R Each Raman peak in is divided by I′ max , obtain the Raman parameter set I′ of tobacco samples op .

[0073] In some embodiments, the number of tobacco samples from each known origin is greater than or equal to five.

[0074] In some embodiments, in step S10 and / or step S100, the Raman spectrometer used includes a Fourier transform Raman spectrometer.

[0075] In some embodiments, in step S10 and / or step S100, the wavelength of the laser of the Raman spectrometer is selected to be 785 nm or 1064 nm.

[0076] In a preferred or optional embodiment, in step S10 and / or step S100, the wavelength of the laser of the Raman spectrometer is selected to be 1064 nm.

[0077] The 1064nm laser source can effectively reduce the fluorescence characteristics of tobacco, making the interference caused by the fluorescence characteristics in tobacco smaller.

[0078] In some embodiments, in step S10 and / or step S100, the laser power of the laser of the Raman spectrometer is selected to be 10 mW to 150 mW.

[0079] The stronger the laser power, the stronger the Raman signal obtained. However, excessive laser power may cause the sample to be damaged by the laser. Selecting a laser power of 10mW to 150mW can obtain a Raman spectrum with a high signal-to-noise ratio without damaging the sample.

[0080] In some embodiments, in step S10 and / or step S100, the number of test spectrum superpositions of the Raman spectrometer is set to 300 times or more.

[0081] Since the signal intensity of tobacco Raman spectrum is weak, multiple scans are required to superimpose the signal and eliminate noise interference to enhance the signal intensity and improve the signal-to-noise ratio. Therefore, based on the characteristics of tobacco samples, the number of superpositions must be greater than or equal to 300 times.

[0082] In some embodiments, step S10 and / or step S100 include: selecting more than eight position points of the tobacco sample to be tested, using a Raman spectrometer to detect the more than eight position points, obtaining more than eight groups of values, and obtaining a Raman spectrum curve regarding Raman shift-Raman intensity based on the average value of the more than eight groups of values.

[0083] The structural uniformity of tobacco samples, the homogeneity and standard deviation of Raman spectral signals are affected by the number of tests. When the number of test points is more than eight, the random error of the test can be lower.

[0084] In some preferred or optional embodiments, the method for identifying tobacco origin based on Raman spectroscopy technology comprises the following steps:

[0085] S1: Select multiple tobacco samples from different origins, with the number of tobacco samples from each origin being no less than five, and use a Fourier transform Raman spectrometer to detect them, select the laser wavelength of the Fourier transform Raman spectrometer to be 1064nm, the laser power to be 10mW-150mW, set the test spectrum superposition times of the Fourier transform Raman spectrometer to be 300 times or more, select more than eight position points of the tobacco samples, use the Raman spectrometer to detect the more than eight position points, obtain more than eight groups of values, and obtain the Raman spectrum curve of the tobacco samples of known origin about Raman shift-Raman intensity according to the average value of the more than eight groups of values;

[0086] S2: Analyze and process the Raman spectrum data obtained in step S1. Select the Raman shift of 2700 cm -1 ~3500cm-1 The Raman spectrum curve segment is connected by connecting the two end points of the Raman spectrum curve segment to obtain a straight line segment, and multiple Raman shifts are selected on the Raman spectrum curve segment. The Raman intensity corresponding to the multiple Raman shifts is subtracted from the Raman intensity corresponding to the corresponding Raman shift on the straight line segment to obtain the Raman peak number set I′ R , to eliminate the fluorescence effect of tobacco itself, and then take the maximum peak I′ on the Raman spectrum curve segment max , the Raman peak number set I′ R Each Raman peak in is divided by I′ max , obtain the Raman parameter set I′ of tobacco samples op .

[0087] For example, I′ opi =I′ Ri / I′ max ;

[0088] Among them, I′ opi is the Raman parameter set I′ of tobacco samples op The ith Raman parameter in Ri is the Raman peak number set I′ R The i-th Raman peak in .

[0089] The number of Raman shifts selected on the Raman spectrum curve segment is n, and the value of i is [1, n].

[0090] S3: Establish tobacco origin and Raman parameter data set I′ op Analyze the Raman spectrum parameters of tobacco from different origins obtained in step S2, classify the Raman parameters of tobacco from the same origin into one category, establish a mapping relationship between the origin and the Raman parameters, and thus establish a database that can identify the origin of tobacco based on the Raman parameters of tobacco;

[0091] Among them, by gathering the Raman parameters of tobacco samples from the same origin together, a Raman parameter data set contained in a certain origin can be obtained. Each Raman parameter in the database corresponds to the origin, forming a mapping relationship. Since tobacco from the same origin is also different, the Raman parameters of different tobaccos from the same origin will also be different.

[0092] S4: Detect the tobacco sample to be tested according to the method provided in step S1 to obtain a Raman spectrum curve of the tobacco sample to be tested about Raman shift-Raman intensity, and obtain a Raman parameter number set I of the tobacco sample to be tested according to the method provided in step S2. op ;

[0093] S5: Calculate the Raman parameter set I of the tobacco sample to be tested op The Raman parameter set I′ of tobacco samples from known origins in the databaseop The squared Euclidean distance between them is given by:

[0094]

[0095] Obtain a×b D values;

[0096] Wherein, a represents the number of known origins, b represents the number of tobacco samples in each known origin; n represents the Raman parameter set I of the tobacco sample to be tested. op The number of Raman parameters in is also the number of Raman parameters of tobacco samples of known origin I′ op The number of Raman parameters in I opi Represents the Raman parameter set I of the tobacco sample to be tested op The i-th Raman parameter in opi The Raman parameter set I′ represents the tobacco sample with known origin op The i-th Raman parameter in .

[0097] That is to say, the tobacco sample to be tested is compared and calculated with all tobacco samples from one of the known origins, and then the tobacco sample to be tested is compared and calculated with all tobacco samples from another known origin, until the comparison and calculation of the tobacco sample to be tested with all tobacco samples from all known origins is completed, and a×b D values ​​are obtained.

[0098] S6: sorting the a×b D values ​​calculated in step S5 in order of size, and classifying the tobacco sample to be tested and the tobacco sample with the smallest D value and known origin into one category, thereby realizing the origin identification of the tobacco sample to be tested.

[0099] The following is a specific embodiment of a method for identifying tobacco origin based on Raman spectroscopy technology. The method for identifying tobacco origin provided by the specific embodiment includes the following steps:

[0100] S11: Nicolet iS50R FT-IR / Raman spectrometer was used, and an InGaAs detector cooled in liquid nitrogen was used to detect and collect Raman signals. The laser wavelength of the spectrometer was 1064nm, the laser power was 50mw, the microscope magnification was 50 times, the test was stacked 300 times, and tobacco samples of known origin were tested to obtain the Raman spectrum curve of the tobacco samples of known origin about Raman shift-Raman intensity;

[0101] S12: Analyze and process the Raman spectrum data obtained in step S11. Select the Raman shift of 2700 cm -1 ~3500cm -1The Raman spectrum curve segment is connected by connecting the two end points of the Raman spectrum curve segment to obtain a straight line segment, and multiple Raman shifts are selected on the Raman spectrum curve segment. The Raman intensity corresponding to the multiple Raman shifts is subtracted from the Raman intensity corresponding to the corresponding Raman shift on the straight line segment to obtain the Raman peak number set I′ R , and then take the Raman shift as 2700cm -1 ~3500cm -1 The maximum peak I′ on the Raman spectrum curve segment max , the Raman peak number set I′ R Each Raman peak in is divided by I′ max , obtain the Raman parameter set I′ of tobacco samples op :

[0102] For example, I′ opi =I′ Ri / I′ max ;

[0103] Among them, I′ opi is the Raman parameter set I′ of tobacco samples op The ith Raman parameter in Ri is the Raman peak number set I′ R The i-th Raman peak in .

[0104] S13: Establish tobacco origin and Raman parameter data set I′ op Analyze the Raman spectrum parameters of tobacco from different origins obtained in step S12, classify the Raman spectrum parameters of tobacco from the same origin into one category, establish a mapping relationship between the origin and the Raman spectrum parameters, and thus establish a database that can identify the origin of tobacco based on the Raman spectrum parameters of tobacco;

[0105] S14: Detect the tobacco sample to be tested according to the method provided in step S11 to obtain a Raman spectrum curve of the tobacco sample to be tested about Raman shift-Raman intensity, and obtain a Raman parameter number set I of the tobacco sample to be tested according to the method provided in step S12. op ;

[0106] S15: Using machine learning methods, calculate the Raman parameter set I of the tobacco sample to be tested op The Raman parameter set I′ of tobacco samples from known origins in the database op The squared Euclidean distance between them is given by:

[0107]

[0108] Obtain a×b D values;

[0109] S16: Establish an association recognition model between the Raman spectrum of tobacco of known origin and the origin, sort the a×b D values ​​calculated in step S15 in order of size, and classify the tobacco sample to be tested and the tobacco sample of known origin with the smallest D value into one category, thereby realizing the origin identification of the tobacco sample to be tested.

[0110] The accuracy of the tobacco origin identification method provided by the embodiment of the present disclosure is verified below.

[0111] For example, the tobacco samples are produced in two categories, A and B. The A sample library contains 27 tobacco samples, and the B sample library contains 30 tobacco samples. According to step S11, a suitable Raman spectrometer and test parameters are selected, and the Raman spectrum data are normalized according to step S12. According to step S13, a Raman spectrum database of known origins is established.

[0112] Among the tobacco samples to be tested that are verified using the tobacco origin identification method provided by the embodiment of the present disclosure, for example, there are 11 tobacco samples originating from A and 11 tobacco samples originating from B.

[0113] For one of the tobacco samples from origin A, select a suitable Raman spectrometer and test parameters according to step S11, and perform normalization on the Raman spectrum data according to step S14. Take the second-order Raman spectrum data, i.e., the Raman shift is 2700 cm -1 ~3500cm -1 The Raman spectrum curve segment data is used to calculate the Raman parameter set I of the tobacco sample to be tested. op The Raman parameter set I′ of tobacco samples from known origins in the database op The squared Euclidean distance between them is shown in the following table.

[0114] For one of the tobacco samples from origin A, the squared Euclidean distances between the Raman parameters of the tested tobacco sample and the sample library sample are ranked from small to large as follows:

[0115]

[0116]

[0117] It can be seen from the results in the table that the origin of the tobacco sample to be tested detected by the tobacco origin identification method provided by the embodiment of the present disclosure is A, and its actual origin is also A.

[0118] The other 21 tobacco samples to be tested were tested for the accuracy of the tobacco origin identification method provided in the embodiment of the present disclosure according to the above method.

[0119] As the input data of the KNN algorithm, when the K value is 3 (that is, in the case of two known origins, three minimum values ​​are selected), the accuracy of tobacco origin classification can reach 95.5%.

[0120] Based on the above-mentioned embodiments of the present disclosure, in the absence of explicit negation or conflict, the technical features of one embodiment may be beneficially combined with one or more other embodiments.

[0121] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A method for identifying tobacco origin, characterized in that: The following steps are involved: S10: testing the tobacco sample to be tested by using a Raman spectrometer to obtain a Raman spectrum curve of Raman shift-Raman intensity; S20: selecting a Raman spectrum curve segment that can characterize the origin of tobacco from the Raman spectrum curve; S30: connecting two end points of the Raman spectrum curve segment to obtain a straight line segment; S40: Obtaining the Raman parameter set I of the tobacco sample to be tested according to the difference information of the Raman intensity between the point set on the Raman spectrum curve segment and the corresponding point set on the straight line segment op ; S50: The Raman parameter set I of the tobacco sample to be tested op The Raman parameter sets I′ of tobacco samples with known origins are op Compare one by one, and select the origin corresponding to the tobacco sample of known origin that is closest to the Raman parameter number set of the tobacco sample to be tested as the origin of the tobacco sample to be tested; The step S40 comprises: Select multiple Raman shifts on the Raman spectrum curve segment, and subtract the Raman intensity corresponding to the Raman shift on the straight line segment from the Raman intensity corresponding to the multiple Raman shifts to obtain the Raman peak number set I R ; Select the maximum peak I on the Raman spectrum curve segment max ; The Raman peak number set I R Each Raman peak in is divided by I max , obtain the Raman parameter set I of the tobacco sample to be tested op .

2. The tobacco origin identification method according to claim 1, characterized in that: In step S20, the Raman shift corresponding to the Raman spectrum curve segment selected to characterize the origin of tobacco is 2700 cm -1 ~3500cm -1 .

3. The tobacco origin identification method according to claim 1 or 2, characterized in that: In step S50, the Raman parameter set I of the tobacco sample to be tested is op The Raman parameter sets I′ of tobacco samples with known origins are op Compare them one by one, including: According to the formula: Calculate to obtain a×b D values, and sort the a×b D values ​​in order of size; Where a represents the number of known origins, b represents the number of tobacco samples in each known origin, and n represents the Raman parameter set I of the tobacco samples to be tested. op The number of Raman parameters in I opi Represents the Raman parameter set I of the tobacco sample to be tested op The ith Raman parameter in opi The Raman parameter set I′ represents the tobacco sample with known origin op The i-th Raman parameter in; D is the Raman parameter set I of the tobacco sample to be tested op Raman parameter set I′ of tobacco samples with known origins op The squared Euclidean distance between .

4. The tobacco origin identification method according to claim 3, characterized in that: In the step S50, selecting the origin corresponding to the tobacco sample of known origin that is closest to the Raman parameter number set of the sample to be tested as the origin of the tobacco sample to be tested includes: The smallest D value is selected from a×b D values, and the tobacco sample of known origin corresponding to the smallest D value is determined. The origin corresponding to the tobacco sample of known origin is the origin of the tobacco sample to be tested.

5. The tobacco origin identification method according to claim 3, characterized in that: Select m smallest D values ​​from a×b D values, where m>a and m is an odd number. The known origin corresponding to the m smallest D values ​​has the largest number of tobacco samples, and this known origin is the origin of the tobacco samples to be tested.

6. The tobacco origin identification method according to claim 3, characterized in that: The value of b is greater than or equal to five.

7. The tobacco origin identification method according to claim 1, characterized in that: In step S50, the Raman parameter set I′ of at least one tobacco sample among a plurality of tobacco samples of known origin op The methods for obtaining include: S100: testing the tobacco sample using a Raman spectrometer to obtain a Raman spectrum curve of Raman shift-Raman intensity; S200: selecting a Raman spectrum curve segment that can characterize the origin of tobacco from the Raman spectrum curve; S300: connecting two end points of the Raman spectrum curve segment to obtain a straight line segment; S400: Obtaining the Raman parameter set I′ of the tobacco sample according to the difference information of the Raman intensity between the point set on the Raman spectrum curve segment and the corresponding point set on the straight line segment op .

8. The tobacco origin identification method according to claim 1, characterized in that: In the step S10, the Raman spectrometer used includes a Fourier transform Raman spectrometer; and / or the wavelength of the laser of the Raman spectrometer is selected to be 785nm or 1064nm; and / or the laser power of the laser of the Raman spectrometer is selected to be 10mW to 150mW; and / or the test spectrum superposition times of the Raman spectrometer are set to 300 times or more.

9. The tobacco origin identification method according to claim 1, characterized in that: The step S10 includes: selecting more than eight position points of the tobacco sample to be tested, using a Raman spectrometer to detect the more than eight position points, obtaining more than eight groups of values, and obtaining a Raman spectrum curve about Raman shift-Raman intensity according to the average value of the more than eight groups of values.

Images