CN105717066B - A kind of near infrared spectrum identification model based on weighted correlation coefficient - Google Patents
A kind of near infrared spectrum identification model based on weighted correlation coefficient Download PDFInfo
- Publication number
- CN105717066B CN105717066B CN201610064824.4A CN201610064824A CN105717066B CN 105717066 B CN105717066 B CN 105717066B CN 201610064824 A CN201610064824 A CN 201610064824A CN 105717066 B CN105717066 B CN 105717066B
- Authority
- CN
- China
- Prior art keywords
- spectrum
- weighted correlation
- near infrared
- identification
- correlation coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000002329 infrared spectrum Methods 0.000 title claims abstract description 26
- 238000001228 spectrum Methods 0.000 claims abstract description 68
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 241000208125 Nicotiana Species 0.000 claims description 18
- 235000002637 Nicotiana tabacum Nutrition 0.000 claims description 18
- 230000003595 spectral effect Effects 0.000 claims description 17
- 239000013598 vector Substances 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 9
- 239000000843 powder Substances 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 235000019504 cigarettes Nutrition 0.000 abstract description 22
- 238000004458 analytical method Methods 0.000 abstract description 14
- 230000007613 environmental effect Effects 0.000 abstract description 3
- 238000004587 chromatography analysis Methods 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 238000001819 mass spectrum Methods 0.000 abstract 1
- 239000000047 product Substances 0.000 description 52
- 238000000034 method Methods 0.000 description 17
- 238000010521 absorption reaction Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000012937 correction Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000001035 drying Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000004445 quantitative analysis Methods 0.000 description 2
- 238000007873 sieving Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000000862 absorption spectrum Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000013401 experimental design Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000004949 mass spectrometry Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000491 multivariate analysis Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012360 testing method Methods 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
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The present invention establishes a kind of near infrared spectrum identification model, and a series of spectrum of normal similar products is scanned using near infrared spectrometer, calculates averaged spectrum and is used as with reference to spectrum；Then the weighted correlation coefficient for calculating product spectrum and reference spectra identifies section by the average value and standard deviation calculation of weighted correlation coefficient, establishes identification model.Compared with the Instrumental Analysis such as traditional chromatography, mass spectrum, there is green, environmental protection, it is simple and fast, easily operated advantage, and institute's established model identification accuracy is high, detection efficiency is high, it is at low cost, provide technical guarantee for the product stability analysis of cigarette enterprise and truth identification.
Description
Technical Field
The invention relates to the technical field of near infrared spectrum, in particular to a near infrared spectrum recognition model based on weighted correlation coefficients, which can be used for authenticity identification and quality stability analysis of cigarette products.
Background
The near infrared spectrum technology has the modern analysis characteristics of high efficiency, green and environmental protection in the analysis process, so that the near infrared spectrum technology becomes one of the spectral analysis technologies which are developed quickly and attract attention in recent years. According to the American Society for Testing and Materials (ASTM), the wavelength range is 780-2526 mn. The absorption of molecules in an NIR area mainly comprises the complex frequency absorption and the frequency doubling absorption of groups such as C-H, 0-H, N-H, C-0 and the like, the absorption intensity of the area is low, the spectral bands are complex and overlap seriously, a classical qualitative and quantitative method cannot be used, calibration modeling must be carried out by means of multivariate statistics, curve fitting, cluster analysis and other methods in chemometrics, and a proper model is combined to realize rapid multicomponent analysis. The near infrared spectrum technology has the advantages of rapidness, no damage, real-time detection and the like, and has become a powerful tool for analyzing industrial products. However, the near infrared spectrum is poor in characteristics, large in data volume, and difficult to obtain reliable results through visual recognition and traditional matching algorithms. Therefore, the development of an effective, fast and highly automated recognition algorithm is urgently needed.
For a long time, the internal quality characterization of cigarette products mainly adopts a sensory evaluation method, and an intuitive and visual quantitative description method is lacked. With the increasingly fierce market competition of products and the increasing production automation degree of industrial enterprises, the stability and control of the quality of the products become more and more prominent, and a rapid, efficient and simple analysis method for evaluating and controlling the quality stability of the cigarette products is urgently needed.
The invention provides an identification model based on a weighted correlation coefficient aiming at the problems. The model is based on near infrared spectra of a series of normal similar products, and an identification model is established through weighting correlation coefficients. In the application of the model, the near infrared spectrum of the similar product which is judged to be normal can be added into the model, so that the identification model is updated, the adaptability is stronger, and the result is more accurate.
According to the invention, the similarity of the spectra is measured through the weighted correlation coefficient, and the weighted correlation coefficient can effectively use the spectral characteristics for similarity calculation, so that the reliability of the result is improved. The near infrared spectrum of the normal similar product is highly similar to the reference spectrum, but is different; not only embodies the common characteristics, but also reflects the individual difference. Therefore, an interval of spectrum similarity is determined, products in the interval are similar products, and otherwise, the products are non-similar products or abnormal products.
Disclosure of Invention
In order to further improve the identification precision of the model, the invention provides a method for extracting the characteristic spectrum based on the weighted correlation coefficient. The identification model provided by the invention consists of a reference spectrum and an identification interval.
The invention uses near-infrared spectrometer to scan the spectrum of a series of normal similar products, then extracts the characteristic data of each spectrum, and uses the average spectrum as the reference spectrum; and calculating a weighted correlation coefficient of the product spectrum and the reference spectrum, calculating an identification interval through the average value and the standard deviation of the weighted correlation coefficient, and establishing an identification model, wherein the identification model consists of the reference spectrum and the identification interval.
The concrete modeling steps are as follows:
(1) firstly, preprocessing the same kind of samples produced in m different batches, scanning the spectrum of the sample by adopting near infrared, and taking a vector siSpectral data (i ═ 1, 2.. times, m) for the ith sample, each spectrum containing n data points.
(2) By the spectral vector siFor the row vectors, a data matrix S of the form is organized,
the average spectrum was calculated by the following formula
(3) Calculate all spectra s andwherein w is the weight wcc
Wherein s isjAndrespectively represent spectra s andthe jth data point of (1); w is ajThe jth data point of the weight vector w.
(4) The weight vector w in step 3 is calculated by the following formula
Wherein the vector d is calculated by the following formula
Wherein,the superscript T denotes matrix transposition.
(5) Calculate wcc mean and standard deviation, respectivelyAnd d. Establishing an identification intervalWhere k is a scaling factor and is set based on the calibration set data such that all wcc are greater thanWill be provided with
As the identification interval of the product.
For unknown products, the near infrared spectrum is first scanned by sxExpressing, the weighted correlation coefficient wcc is then calculated by the formula in step 3x. Decision wccxWhether it is in the identification interval If yes, the unknown product is considered to be the same as the correction set product; will sxAdding the identification interval into a correction set, repeating the steps 1-4, and updating the identification interval. If not, the unknown product is considered to be different from the correction set product.
Calculating the mean value and standard deviation of the weighted correlation coefficients wcc of all the spectra of the same kind by using a weighted correlation coefficient calculation formula, wherein the mean value is used as the mean valueIs expressed by d, standard deviation is expressed by d, and identification interval is establishedWhere k is the scaling factor.
The identification interval and the calibration set data are calculated according to the weighted correlation coefficient, and the weighted correlation coefficient wcc of all the same type of spectra is larger than that of the same type of spectraThe identification interval of the products is
When the model is applied, byScanning the spectrum of the sample to be analyzed and calculating a weighted correlation coefficient wccxIf the coefficient falls within the identification intervalCan be judged to be the same kind of normal products.
The model is based on near infrared spectra of a series of normal similar products, and an identification model is established through weighting correlation coefficients.
The near infrared spectrum identification model based on the weighted correlation coefficient comprises the following steps: comprises the steps of crushing a sample into 40-80 meshes before scanning. The sample is cut tobacco, tobacco stem or tobacco powder.
In the application of the model, the near infrared spectrum of the similar product which is judged to be normal can be added into the model, so that the supplement and the update of the identification model are realized, the adaptability of the model is stronger, and the prediction result is more accurate.
Compared with the prior art, the invention has the following remarkable advantages:
1. according to the method for establishing the product authenticity and stability identification model based on the weighted correlation coefficient, the weighted correlation coefficient can effectively use the spectral characteristics for similarity calculation, and the reliability of the identification model is greatly improved.
2. An interval of spectrum similarity is determined by weighting the correlation coefficients, so that the common characteristics of each spectrum can be reflected, and the individual difference is reflected. The near infrared spectrum of a normal similar product is highly similar to the reference spectrum, an area space is established, the sample scanning spectrum falls in the interval to be the similar product, and non-similar products or abnormal products are outside the interval, so that the phenomenon of inaccurate judgment can be effectively avoided, the model identification precision is improved, and the technical support is provided for the quality stability analysis and the authenticity identification of products of cigarette manufacturing enterprises.
3. Compared with the traditional instrument analysis such as chromatography and mass spectrometry, the near infrared spectrum technology does not use chemical reagents in the whole analysis process, has the advantages of environmental protection, simplicity, rapidness and easy operation, applies chemical metering tools such as matrixes, weighted correlation coefficients and the like in model establishment, and has high identification accuracy, high detection efficiency and low cost of the established model.
Description of the drawings:
FIG. 1 is a modeling flow diagram of the present invention;
FIG. 2 is a near-infrared scanning original spectrogram of cut tobacco of cigarette;
FIG. 3 is an identification model established by the near infrared spectrum of the cut tobacco of the A-brand cigarette;
FIG. 4 is an A, B brand classification recognition model;
the specific implementation mode is as follows:
the following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
The modeling process of the embodiment of the invention is as follows: firstly, carrying out experimental design, carrying out sample collection according to the design, preprocessing a representative collected sample, carrying out spectrum collection by using a near-infrared spectrometer, and optimizing collected spectrum parameters, wherein the spectrum preprocessing method adopts Norris derivative smoothing filtering, differential processing, multivariate scattering correction, standard normalization and other methods; the band selection optimizes the spectral band by using the modes of partial least square method, genetic algorithm, non-information variable elimination and the like. After spectrum optimization, a qualitative analysis model is established, a near-infrared identification model is established according to the steps of extracting spectral characteristic data, establishing a data matrix, calculating an average spectrum, calculating a weighted correlation coefficient and the like, after the model is established, spectrum scanning is carried out on a sample to be tested, and the model is applied for analysis. See FIG. 1
Example one
In the embodiment, the identification model is used for stability identification of the same kind of products.
1. Laboratory apparatus
MPA type Fourier near infrared spectrometer produced by BRUKER company (Germany) and 1095Cyclotec (XF-98B) type cyclone precision pulverizer.
2. Sample collection
In order to ensure that the established identification model has wide applicability in predicting cigarette products produced by different machines in different time periods, 83 normal samples of the cigarette products of the same brand (A) produced by 3 different machines in 2015 in 1-12 months are selected for model establishment in the experimental sample, and 31 unknown samples are selected for model external verification.
3. Sample preparation
Stripping cut tobacco of cigarette, drying in a drying oven at 40 ℃ to keep the moisture of the sample basically consistent, fully crushing by a 1095Cyclotec (XF-98B) type cyclone precision crusher, and sieving by a 60-mesh sieve.
4. Spectral scanning and data processing
The method comprises the following steps of scanning the spectrogram of a cigarette tobacco sample by using an MPA type Fourier near infrared spectrometer (with a near infrared quantitative analysis diffuse reflection gold-plated large integrating sphere and a sample rotator sampling accessory) produced by BRUKER company (Germany), and processing the spectrogram by using qualitative analysis software QUANT6.5 in Bruker OPUS, wherein the method comprises the following specific operations: the tobacco powder was loaded into a sample cup with a height of about 3cm, the weight was pressed against the sample for 10s and removed, the quartz glass at the bottom of the cup was wiped clean with gauze, and the sample cup was then placed on a rotating platform for NIR scanning. The operating parameters are as follows: the spectrum scanning range is 12000-4000 cm-1Spectral resolution of 8cm-1The number of scans was 64 times (about 30S). Spectral data is collected in a transmission mode and processed into a first order differential of the absorption spectrum. The original scanning image of the cut tobacco of the cigarette is shown in figure 2. In the modeling process, to eliminate the influence of noise and baseline, first-order derivative 9-point smoothing (Sa) is adoptedvitzky-Golay) pre-processes the scanned raw spectra. After the sample was scanned, the spectral data was processed with statistical software.
5. Identification model building
The model establishment steps are as follows:
5.1 by spectral vector siFor the row vectors, a data matrix S of the form is organized,
the average spectrum was calculated by the following formula
By the above formula, calculate
5.2 calculate all spectra s andwherein w is the weight wcc
Wherein s isjAndrespectively represent spectra s andthe jth data point of (1); w is ajThe jth data point of the weight vector w.
5.3 the weight vector w in the above step is calculated by the following formula
Wherein the vector d is calculated by the following formula
Wherein,the superscript T denotes matrix transposition.
5.4 calculate the mean and standard deviation of wcc, expressed asAnd d. Establishing an identification intervalWhere k is a scaling factor and is set based on the calibration set data such that all wcc are greater thanWill be provided with As the identification interval of the product.
The average spectrum of the spectrums of 83A-grade cigarette tobacco shreds is used as a reference spectrum; calculating the weighting correlation coefficients of the spectra of 83 cut tobaccos of the brand and the reference spectrum through the formula; the mean and standard deviation of these weighted correlation coefficients are 0.9998 and 0.0002025, respectively; through analysis, all the weighted correlation coefficients are larger than 0.9998-3 × 0.0002025-0.9992, so that the identification interval of the product is determined as [0.9992,1], and the identification model established by the near infrared spectrum of the A-brand cut tobacco of cigarettes is shown in fig. 3:
after the same batch of samples are pretreated, the samples are placed in the air for 6 hours, then the upper level scans the spectrum, the weighted correlation coefficient of the near infrared spectrum of the product and the reference spectrum in the model is calculated, and the calculation result is 0.9985. This value is in the model identification interval [0.9992,1], so the product is determined to be different from the normal product described by the identification model. Actually, the product is placed in the air before near infrared spectrum scanning, moisture absorption is serious, although the nature of tobacco shreds is unchanged, moisture is changed, quality is affected, no abnormal sample can be seen from the spectrum, but the weighted correlation coefficient of the scanned spectrum is far lower than that of the similar product, and the model can be used for quality stability analysis of the same product.
6. Model validation
In order to better verify the identification capability of the model, the experiment adopts an external verification method, 31 batches of samples which do not participate in modeling are selected, the model is used for identifying A-brand cigarettes produced by different batches and different machines, and the result is shown in table 1:
TABLE 1 recognition results of brand "A" product feature model
The results show that: 31 samples of different batches and different machines are successfully identified, the identification rate is 100%, and the established model is high in prediction accuracy and can be used for quality stability analysis of cigarette products.
Example two
In the present embodiment, the recognition model is used for authenticity recognition of a product.
1. Laboratory apparatus
MPA type Fourier near infrared spectrometer produced by BRUKER company (Germany) and 1095Cyclotec (XF-98B) type cyclone precision pulverizer.
2. Sample collection
The cigarette samples selected in the embodiment are of the brand A and the brand B, 5 different machines are selected for the product, the production time is 2015 for 1-12 months, 83 normal samples of the brand A are selected for model building, and 17 unknown samples of the brand B are selected for model authenticity identification.
3. Sample preparation
Stripping off cut tobacco of cigarette, drying in a 40 ℃ oven to keep the moisture of the sample basically consistent, fully crushing by a 1095Cyclotec (XF-98B) type cyclone precision crusher, and sieving by a 80-mesh sieve.
The method for spectrum scanning, data processing and model establishment in this embodiment is the same as that in the first embodiment, and the established A, B brand classification and identification model is shown in fig. 4:
the near infrared spectrum of 51 non-similar products with the brand number of B cut tobacco is scanned by adopting the near infrared, the weighted correlation coefficient of the reference spectrum in the identification model is calculated, and the result is shown as a solid point in figure 2. As can be seen from the figure, all the data points are located outside the identification interval, and the identification rate is 100%, so that the products are judged to be non-homogeneous products. The conclusion is completely consistent with the actual situation, thereby proving the effectiveness of the recognition model.
4. Model validation
In order to better verify the identification capability of the model, the experiment adopts an external verification method, 29 batches of cigarette samples with different brands which do not participate in modeling are selected, the model is used for identifying the fake cigarette with brand A and the cigarettes with brand B and brand C collected in the market, and the specific results are shown in table 1:
TABLE 1 recognition results of brand "A" product feature model
The results show that: 29 samples of cigarettes which are not of the A-grade are successfully identified, the identification rate is 100%, and the established model can be used for identifying the authenticity of products.
The above-mentioned embodiments of the present invention are merely examples for clearly illustrating the invention, and are not intended to limit the embodiments of the present invention, and it will be apparent to those skilled in the art that other variations and modifications can be made on the basis of the above description.
Claims (6)
1. A near infrared spectrum recognition model based on a weighted correlation coefficient utilizes a near infrared spectrometer to scan the spectrums of a series of normal similar products, then extracts the characteristic data of each spectrum, and takes an average spectrum as a reference spectrum; calculating a weighted correlation coefficient of the product spectrum and the reference spectrum, calculating an identification interval through the average value and the standard deviation of the weighted correlation coefficient, and establishing an identification model, wherein the identification model consists of the reference spectrum and the identification interval;
the model establishment comprises the following steps:
(1) spectral scanning: performing near infrared spectrum scanning on a sample to be detected, and extracting a characteristic spectrum;
(2) establishing a data matrix: by a vector siSpectral data (i ═ 1, 2.. times, m) representing the ith sample, each spectrum containing n data points; by the spectral vector siAs row vectors, the spectral data matrix S is of the form,
(3) calculating the average spectrum: calculating on the basis of the matrix formula in the step (2) as follows:
in the formula,
(4) calculate all spectra s andis expressed as wcc, and the formula is as follows:
in the formula, sjAndthe jth data point representing spectra s and s, respectively; w is ajRepresents the jth data point of the weight vector w, w being the weight;
(5) the weight vector w in step (4) is calculated as follows:
in the formula, the vector d is calculated as follows:
in the formula,the superscript T denotes matrix transposition.
2. The near infrared spectral recognition model based on weighted correlation coefficients of claim 1, wherein: calculating the mean value and standard deviation of the weighted correlation coefficients wcc of all the spectra of the same kind by using a weighted correlation coefficient calculation formula, wherein the mean value is used as the mean valueIs expressed by d, standard deviation is expressed by d, and identification interval is establishedWhere k is the scaling factor.
3. The near infrared spectrum recognition model based on weighted correlation coefficients of claim 1 or 2, wherein: the identification interval and the calibration set data are calculated according to the weighted correlation coefficient, and the weighted correlation coefficient wcc of all the same type of spectra is larger than that of the same type of spectraThe identification interval of the products is
4. The near infrared spectral recognition model based on weighted correlation coefficients of claim 3, wherein: in applying the model, the weighted correlation coefficients wcc are calculated by scanning the spectrum of the sample to be analyzedxIf the coefficient falls within the identification intervalCan be judged to be the same kind of normal products.
5. The near infrared spectral recognition model based on weighted correlation coefficients of claim 4, wherein: comprises the steps of crushing a sample into 40-80 meshes before scanning.
6. The near infrared spectral recognition model based on weighted correlation coefficients of claim 5, wherein: the sample is cut tobacco, tobacco stem or tobacco powder.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610064824.4A CN105717066B (en) | 2016-01-29 | 2016-01-29 | A kind of near infrared spectrum identification model based on weighted correlation coefficient |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610064824.4A CN105717066B (en) | 2016-01-29 | 2016-01-29 | A kind of near infrared spectrum identification model based on weighted correlation coefficient |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105717066A CN105717066A (en) | 2016-06-29 |
| CN105717066B true CN105717066B (en) | 2018-09-18 |
Family
ID=56154383
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610064824.4A Active CN105717066B (en) | 2016-01-29 | 2016-01-29 | A kind of near infrared spectrum identification model based on weighted correlation coefficient |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105717066B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107402192A (en) * | 2017-03-03 | 2017-11-28 | 广西中烟工业有限责任公司 | A kind of method of quick analysis essence and flavoring agent quality stability |
| CN108982402B (en) * | 2018-08-01 | 2021-01-08 | 华北电力科学研究院有限责任公司 | Method and system for predicting harm of similar organic pollutants to environment |
| CN109738342B (en) * | 2019-03-18 | 2022-02-18 | 山东金璋隆祥智能科技有限责任公司 | Method for detecting particle size distribution based on near infrared spectrum technology |
| CN110632024B (en) * | 2019-10-29 | 2022-06-24 | 五邑大学 | Quantitative analysis method, device and equipment based on infrared spectrum and storage medium |
| CN112834451B (en) * | 2021-01-12 | 2023-04-18 | 深圳网联光仪科技有限公司 | Sample identification method and device based on infrared spectrum and storage medium |
| CN112697744A (en) * | 2021-01-14 | 2021-04-23 | 中国林业科学研究院木材工业研究所 | Infrared spectrum-based identification method for Dongfei yellow sandalwood artware |
| US20220397560A1 (en) * | 2021-06-10 | 2022-12-15 | Thermo Finnigan Llc | Auto outlier injection identification |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5661558A (en) * | 1994-06-30 | 1997-08-26 | Hitachi, Ltd. | Optical detector for flowing sample and optical detection method for flowing sample |
| CN101620180A (en) * | 2009-05-08 | 2010-01-06 | 合肥美亚光电技术有限责任公司 | Method for rapidly detecting tea quality through near infrared technology |
| CN105891147A (en) * | 2016-03-30 | 2016-08-24 | 浙江中烟工业有限责任公司 | Near infrared spectrum information extraction method based on canonical correlation coefficients |
-
2016
- 2016-01-29 CN CN201610064824.4A patent/CN105717066B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5661558A (en) * | 1994-06-30 | 1997-08-26 | Hitachi, Ltd. | Optical detector for flowing sample and optical detection method for flowing sample |
| CN101620180A (en) * | 2009-05-08 | 2010-01-06 | 合肥美亚光电技术有限责任公司 | Method for rapidly detecting tea quality through near infrared technology |
| CN105891147A (en) * | 2016-03-30 | 2016-08-24 | 浙江中烟工业有限责任公司 | Near infrared spectrum information extraction method based on canonical correlation coefficients |
Non-Patent Citations (3)
| Title |
|---|
| 近红外光谱峰谷相关系数法的研究与验证初探;张学博 等;《中国药师》;20131231;第16卷(第10期);1540－1544 * |
| 近红外光谱相关系数法测定野木瓜片中异性有机物;徐东来 等;《药物分析杂志》;20111231;第31卷(第7期);1423－1424 * |
| 近红外特征谱段相关系数法测定中药胶囊中添加枸橼酸西地那非;王学良;《分析化学》;20091231;第37卷(第12期);1825－1828 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105717066A (en) | 2016-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105717066B (en) | A kind of near infrared spectrum identification model based on weighted correlation coefficient | |
| WO2018010352A1 (en) | Qualitative and quantitative combined method for constructing near infrared quantitative model | |
| CN108181263B (en) | Tobacco leaf position feature extraction and discrimination method based on near infrared spectrum | |
| CN106501208A (en) | A kind of tobacco style similitude sorting technique based near infrared light spectrum signature | |
| CN109115692B (en) | Spectral data analysis method and device | |
| CN104215591A (en) | Damage-free visible-near infrared light spectrum detecting method | |
| CN114088661B (en) | Tobacco leaf baking process chemical composition online prediction method based on transfer learning and near infrared spectrum | |
| CN106918572A (en) | The assay method of potato content in potato compounding staple food | |
| CN109540837B (en) | Method for rapidly detecting lignocellulose content of ramie leaves by near infrared | |
| CN111257277B (en) | Tobacco leaf similarity judgment method based on near infrared spectrum technology | |
| WO2020248961A1 (en) | Method for selecting spectral wavenumber without reference value | |
| CN109358022A (en) | A kind of method of the quick-fried pearl type of quick discrimination cigarette | |
| CN103743705A (en) | Rapid detection method for sorghum halepense and similar species | |
| CN113655027B (en) | Method for near infrared rapid detection of tannin content in plants | |
| CN105223140A (en) | The method for quickly identifying of homology material | |
| CN104316492A (en) | Method for near-infrared spectrum measurement of protein content in potato tuber | |
| Liu et al. | Research on the online rapid sensing method of moisture content in famous green tea spreading | |
| CN110887809B (en) | Method for measuring stem content in tobacco shreds based on near infrared spectrum technology | |
| CN113030007B (en) | Method for rapidly testing quality stability of tobacco essence based on similarity learning algorithm | |
| CN114295579A (en) | Tobacco mellowing determination method based on near infrared spectrum | |
| CN110186870B (en) | Method for distinguishing fresh tea leaf producing area of Enshi Yulu tea by extreme learning machine spectrum model | |
| CN113607683A (en) | Automatic modeling method for near infrared spectrum quantitative analysis | |
| CN102706811A (en) | Method for identifying quality of sugar material of cigarette by applying near infrared spectrum analysis technology | |
| CN110596038A (en) | Method for rapidly determining starch content of sweet potatoes | |
| CN117589708A (en) | Establishment method and application method of protein database based on near infrared rapid detection of tissue protein |
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 |