CN106404748B - A kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition method - Google Patents
A kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition method Download PDFInfo
- Publication number
- CN106404748B CN106404748B CN201610802039.4A CN201610802039A CN106404748B CN 106404748 B CN106404748 B CN 106404748B CN 201610802039 A CN201610802039 A CN 201610802039A CN 106404748 B CN106404748 B CN 106404748B
- Authority
- CN
- China
- Prior art keywords
- spectral
- rice
- sample
- identification
- induced breakdown
- 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
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000002536 laser-induced breakdown spectroscopy Methods 0.000 title claims abstract description 41
- 235000013339 cereals Nutrition 0.000 title claims abstract description 39
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 33
- 230000003595 spectral effect Effects 0.000 claims abstract description 123
- 235000007164 Oryza sativa Nutrition 0.000 claims abstract description 112
- 235000009566 rice Nutrition 0.000 claims abstract description 112
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 15
- 238000012706 support-vector machine Methods 0.000 claims abstract description 9
- 238000001228 spectrum Methods 0.000 claims description 37
- 238000012549 training Methods 0.000 claims description 20
- 238000013145 classification model Methods 0.000 claims description 17
- 238000013507 mapping Methods 0.000 claims description 13
- 239000000843 powder Substances 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 9
- 238000002360 preparation method Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 238000012272 crop production Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000013480 data collection Methods 0.000 claims description 2
- 238000011156 evaluation Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 claims description 2
- 240000007594 Oryza sativa Species 0.000 claims 1
- 241000209094 Oryza Species 0.000 abstract description 111
- 241000894007 species Species 0.000 abstract 1
- 239000000523 sample Substances 0.000 description 32
- 238000001514 detection method Methods 0.000 description 17
- 238000005516 engineering process Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 210000002381 plasma Anatomy 0.000 description 9
- 239000000126 substance Substances 0.000 description 9
- 235000013305 food Nutrition 0.000 description 6
- 239000004033 plastic Substances 0.000 description 5
- 229920003023 plastic Polymers 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000000513 principal component analysis Methods 0.000 description 4
- 241001481833 Coryphaena hippurus Species 0.000 description 3
- 241000196324 Embryophyta Species 0.000 description 3
- 240000002853 Nelumbo nucifera Species 0.000 description 3
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 3
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 239000004615 ingredient Substances 0.000 description 3
- 238000004611 spectroscopical analysis Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000005303 weighing Methods 0.000 description 3
- 244000144730 Amygdalus persica Species 0.000 description 2
- 241000282326 Felis catus Species 0.000 description 2
- 241000574138 Ozothamnus diosmifolius Species 0.000 description 2
- 235000006040 Prunus persica var persica Nutrition 0.000 description 2
- 235000021329 brown rice Nutrition 0.000 description 2
- 150000001720 carbohydrates Chemical class 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 239000003205 fragrance Substances 0.000 description 2
- 238000004817 gas chromatography Methods 0.000 description 2
- 229910052739 hydrogen Inorganic materials 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 229910052755 nonmetal Inorganic materials 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 230000001953 sensory effect Effects 0.000 description 2
- 238000010183 spectrum analysis Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 101000694017 Homo sapiens Sodium channel protein type 5 subunit alpha Proteins 0.000 description 1
- 235000010254 Jasminum officinale Nutrition 0.000 description 1
- 240000005385 Jasminum sambac Species 0.000 description 1
- BUGBHKTXTAQXES-UHFFFAOYSA-N Selenium Chemical compound [Se] BUGBHKTXTAQXES-UHFFFAOYSA-N 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007705 chemical test Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000009616 inductively coupled plasma Methods 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004949 mass spectrometry Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 239000002366 mineral element Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000010239 partial least squares discriminant analysis Methods 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 229910052711 selenium Inorganic materials 0.000 description 1
- 239000011669 selenium Substances 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000002470 solid-phase micro-extraction Methods 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 150000003722 vitamin derivatives Chemical class 0.000 description 1
- 238000000209 wet digestion 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
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/71—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light thermally excited
- G01N21/718—Laser microanalysis, i.e. with formation of sample plasma
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N1/00—Sampling; Preparing specimens for investigation
- G01N1/28—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q
- G01N1/286—Preparing specimens for investigation including physical details of (bio-)chemical methods covered elsewhere, e.g. G01N33/50, C12Q involving mechanical work, e.g. chopping, disintegrating, compacting, homogenising
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/02—Food
- G01N33/10—Starch-containing substances, e.g. dough
Landscapes
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Optics & Photonics (AREA)
- Plasma & Fusion (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition methods, this method includes establishing two processes of svm classifier model and Production area recognition, using algorithm of support vector machine, the advantages of species analysis quick in conjunction with LIBS, quick, Accurate classification is carried out to different sources rice, and a plurality of characteristic spectral line of identity element biggish and near one another enhances the otherness between different sources cereal crops (such as rice) using signal-to-background ratio by the way of combining, to improve the accuracy of identification of algorithm.This method combines laser induced breakdown spectroscopy with SVM algorithm, to achieve the purpose that quick and precisely to identify.
Description
Technical Field
The invention belongs to the technical field of spectral analysis, and particularly relates to a method for identifying the production area of cereal crops, particularly rice, by adopting a laser-induced breakdown spectroscopy technology, wherein the production area is also called a source area.
Background
The rice is praised as the head of five cereals, is one of the main food crops in China, occupies about 30 percent of the cultivation area of the food crops, and is basic food for supplementing various trace nutrient elements required by human bodies due to the rich carbohydrate, vitamin and mineral substances. However, the nutritional ingredients of rice vary greatly depending on the variety, place of production, and growth conditions, and in addition, the poor phenomena of "poor quality rice pretends to be good quality rice" and "place of production pretends" are continuously occurring in the market, so that an effective detection method is required for identifying the quality and place of production of rice circulating in the market. Traditional methods of identification mainly include sensory and chemical tests. The sensory detection has strong subjectivity and is time-consuming and labor-consuming; chemical detection requires professional personnel to perform complicated chemical pretreatment on samples, takes long time, and cannot meet the requirements of rapid, green, large-batch and other treatments. Therefore, a fast, environment-friendly and accurate detection method is urgently needed to be researched to realize the identification of the rice.
Laser-induced breakdown spectroscopy (LIBS) is a novel method for component analysis. The technology can rapidly detect samples in large batch; the sampling point has small size and small destructiveness; and can analyze a plurality of elements simultaneously; the detection under severe environment can be realized, and the method is widely applied to the fields of industry, archaeology, soil, water body, food detection and the like.
Research paper "distinguishing rice storage time and production place by gas chromatography in combination with chemometrics" (analysis and test article, volume 32, published in 10 months of 2013) adopts gas chromatography to analyze volatile components of rice samples in different storage times and different production places respectively, and rice samples are classified and discriminantly analyzed by a Principal Component Analysis (PCA) and a partial least squares discriminant analysis (PLS-DA) to obtain the recognition rates of the rice samples in different storage times and different production places respectively of 96% and 100%. In the research paper, research on the technology for identifying geographical indication wuchang rice based on inorganic element analysis (spectroscopy and spectral analysis, 36 th volume, 21 st, 2016, published in 3 months), inductively coupled plasma spectroscopy and mass spectrometry are applied to determine the content of inorganic elements in rice, and wuchang rice and unusual rice are identified by combining PCA, Fisher discrimination and artificial neural network, and the results show that the PCA recognition effect is poor, and the Fisher discrimination and artificial neural network recognition accuracy rates respectively reach 93.5% and 96.4%. The research method obtains higher precision in rice classification, but solid-phase micro-extraction and wet digestion are required to be carried out on rice, so that the detection is complex. Chinese patent document "a method and apparatus for improving the accuracy of laser probe plastic identification" (publication No. CN104730041A, publication date 2015 6-24) discloses a method for classifying plastics based on the LIBS technique of SVM. The recognition rate is improved by increasing the weight of 3 nonmetal characteristic spectral lines and increasing the matrix difference among different plastics. SVM has achieved good results in plastic identification, but has not been applied to agricultural products (such as rice production areas).
From the research, the existing rice classification detection technology mainly uses a chemical detection means, is complex and time-consuming, and cannot meet the requirements of industrial application.
Disclosure of Invention
Aiming at the problems of low efficiency and complex detection process of traditional rice production place identification and detection, the invention provides a method for identifying the production place of cereal crops by using a laser-induced breakdown spectroscopy technology.
The invention provides a multispectral combined laser-induced breakdown spectroscopy grain crop production place identification method which is characterized by comprising two processes of establishing an SVM classification model and identifying a production place, and the method comprises the following steps:
(1) establishing an SVM classification model:
1.1. sample preparation:
preparing powder samples with the same mass into flaky samples with uniform thickness, wherein the samples are n cereal crops with different producing areas;
1.2. collecting spectral data of a sample:
LIBS spectrum data collection is carried out on N grain crops in different producing areas, m plasma spectrums are collected for each grain crop sample in each producing area, N-N-m plasma spectrums are obtained in total, wherein m is the spectrum magnitude number of each sample collected by the LIBS, and m-1, 2, … and 1000;
1.3. selecting characteristic spectral lines
Analyzing plasma spectrums of collected n grain crops in different producing areas, selecting s strongest characteristic spectral lines in each main quantity element as algorithm analysis indexes, reading intensity values of the characteristic spectral lines, selecting one characteristic spectral line intensity in the main quantity element to perform normalization processing on the characteristic spectral intensities of other s-1 main quantity elements, wherein s-1 represents the number of the indexes to be analyzed;
1.4. spectral line combination
Combining the same characteristic spectral lines near the s strongest characteristic spectral lines in the selected principal elements to finally obtain q combined indexes to be analyzed;
1.5.SVM Classification model establishment
Mapping linearly indivisible data into a high-dimensional space in a mode of constructing a kernel mapping function, so as to obtain linear distinction in the high-dimensional space, wherein the kernel mapping function adopts a radial basis function:
selecting partial group spectrums of N groups of cereal crop spectrum data as a training set for establishing a support vector machine model, and using the rest group spectrums as a model test set for testing the identification precision of the established support vector machine model;
(2) origin identification
Firstly, according to the steps from 1.1 to 1.4, carrying out sample preparation, spectral data acquisition, characteristic spectral line selection and spectral line combination on cereal crops to be identified to obtain a plurality of groups of cereal crop spectral data;
then, the spectral data of the unknown samples are classified by using the SVM classification models of the cereal crops in different producing areas established in the step 1.5, so as to obtain corresponding producing area prediction.
The invention adopts a support vector machine algorithm, combines the advantages of LIBS rapid substance analysis, rapidly and accurately classifies the rice in different producing areas, and enhances the difference between cereal crops (such as rice) in different producing areas by adopting a mode of combining a plurality of characteristic spectral lines of the same element with larger signal-to-back ratio and close to each other, thereby improving the identification precision of the algorithm. Specifically, the method of the invention has the following characteristics and effects:
(1) compared with the traditional rice production place identification method, the method does not need complex chemical pretreatment, only simply tabletting and sample preparation are carried out, and the spectral signal acquisition is carried out on the rice in different production places by the laser-induced breakdown spectroscopy substance component analysis technology, so that the sample detection time and the complex chemical analysis process are reduced, the secondary pollution is avoided, and the detection efficiency is improved.
(2) The method of the invention increases the difference between rice in different producing areas by a method of combining the characteristic spectral lines of the same elements with larger signal-to-back ratio and close to each other, but does not increase the number of analysis variables, thereby improving the classification effect of the SVM classifier.
(3) The method has high flexibility, the introduced punishment parameter c of the SVM loose variable and the mapping kernel parameter g can effectively improve the generalization capability of the classifier, and the algorithm has high training efficiency and quick parameter adjustment. The method can be combined with other intelligent algorithms such as Genetic Algorithm (GA), Random Forest (RF), Neural Network (ANN) and the like, so as to further improve the accuracy of classification.
(4) It should be noted that the method of the present invention is not limited to the rice producing area, and can realize the identification of different types of rice in different producing areas. The method is also suitable for identifying the production places of other cereal crops.
Different from the existing method, the accuracy of the SVM classification model is improved through the combination of the characteristic spectral lines near the principal component elements, and the method has the advantages of rapidness, greenness and high identification accuracy.
Drawings
Fig. 1 is a schematic structural diagram of a rice production place identification device with laser-induced breakdown spectroscopy according to an embodiment of the invention:
wherein, 1, laser; 2. a laser wavelength mirror; 3. a focusing lens; 4. a sample to be tested; 5. a signal acquisition device; 6. an optical fiber; 7. a spectrometer; 8. triggering a line; ICCD; 10. a data line; 11. a displacement platform; 12. and (4) a computer.
FIG. 2 is a LIBS spectrum of the golden dolphin Thailand jasmine rice in the wavelength range of 200-900 nm.
FIG. 3 is a comparison of classification using a single characteristic line in combination with multiple characteristic lines.
The specific implementation mode is as follows:
the following further describes embodiments of the present invention with reference to specific examples. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Due to the difference of the growth environment of the rice, the ingredient content of the rice growing from different producing areas is directly different, so that the rice from different producing areas can be distinguished according to the substance ingredients in the rice. The embodiment of the invention provides a method for identifying rice production places by using a multispectral combined support vector machine algorithm assisted with a laser-induced breakdown spectroscopy technology.
The method comprises two processes of establishing an SVM classification model and identifying the origin, and specifically comprises the following steps:
(1) establishing an SVM classification model:
1.1. sample preparation
Firstly, respectively crushing rice in a known production place by using a crusher to prepare uniform powder samples, then weighing the powder samples with the same mass, and preparing the powder samples into flaky samples with uniform thickness by using a tablet press.
1.2. Spectral data acquisition of a sample
In order to obtain the LIBS signal with high spectral intensity and signal-to-noise ratio, rice of one production place of n different production places is selected to carry out optimization on experimental parameters of acquisition delay time, laser energy and defocusing amount of the spectral signal. The spectral intensity stability and signal-to-noise ratio of major elements C, N, H and the like in the rice are used as evaluation indexes, the spectral detection stability index is determined by calculating Relative Standard Deviation (RSD), the specific RSD is (the arithmetic mean of the Standard Deviation/calculation result) 100%, and the smaller the RSD value is, the better the RSD value is; the signal-to-noise ratio is specifically calculated by subtracting the background signal intensity value from the collected spectral intensity value, divided by the noise value, and selecting the signal-to-noise ratio that is the largest. And comprehensively determining the optimal experimental conditions by optimizing the spectral stability and the maximum signal-to-noise ratio of the principal component. Under the best experimental conditions, LIBS spectrum data acquisition is carried out on N rice in different producing areas, m plasma spectrums are acquired for each producing area rice sample, and N-N-m plasma spectrums are obtained in total, wherein N is the number of the collected rice producing areas; m is the LIBS collected spectrum amplitude, and m is 1,2, … and 1000.
1.3. Selecting characteristic spectral lines
The method comprises the steps of analyzing plasma spectrums of collected n rice in different producing areas, selecting the strongest characteristic spectral line of s principal elements as an index to be analyzed by an algorithm, reading the intensity value of each characteristic spectral line by using spectral data processing software, selecting one characteristic spectral line intensity of the s principal elements to perform normalization processing on the characteristic spectral intensities of other s-1 principal elements as sample data, wherein s represents the number of analysis indexes, and s is 5,6, … and 100, and finally obtaining s-1 indexes to be analyzed by uncombined spectral lines.
1.4. Spectral line combination
Combining p same-kind characteristic spectral lines near the strongest characteristic spectral line of the selected s principal elements, and normalizing the characteristic spectral line intensity of the selected uncombined one of the combined s principal elements on the characteristic spectral line intensities of other s-1 principal elements to obtain q combined indexes to be analyzed (the number of the combined indexes to be analyzed q is the same as the number of the uncombined analysis indexes s-1). The value of p is determined by the material composition and its physical properties, and in general, p is 1,2, …, 100.
1.5.SVM Classification model establishment
The mathematical process of training the SVM classification model is as follows: let each instance in the training set be (X)i,yi)，i＝1,…,n，Xi,yiRespectively representing the ith sample data and its corresponding tag value, wherein Xi(xi1,xi2,…,xir) E is equal to Rr, Rr represents a training sample data set, and r represents the number of attribute values; y isiE {1,2,3, … }, 1,2,3, … are label values for each set of attribute values. The important characteristic of the algorithm is that the linear indivisible data is mapped into a high-dimensional space in a mode of constructing a kernel function, so that linear distinction is obtained in the high-dimensional space. The invention selects a Radial Basis Function (RBF) as a kernel mappingFunction of ray:
the resulting nonlinear SVM classifier equation is:
constraint conditions are as follows:
wherein,sample data representing a prediction set;a mapping function representing a prediction set sample;representing a distance of two norms,/, representing the number of training sets, αiIs a lagrange multiplier; b is a bias factor in the equation; c is a penalty parameter, and g is a mapping kernel parameter.
From the equation, it can be seen that the c, g parameters have a significant effect on model building. Generally setting the selection ranges of c and g: c 2e，g＝2w，e、w∈{-10,-9,…,9,10}。
In the embodiment of the invention, the support vector machine software is adopted for data processing, wherein part of the group spectra of the N groups of rice spectrum data are selected as a training set for establishing the support vector machine model, and the rest group spectra are used as a model test set for testing the identification precision of the established support vector machine model.
The Support Vector machine software may be selected from a Support Vector machine software tool box (LIBSVM) developed by people such as fructus alpiniae oxyphyllae.
(2) Origin identification
Firstly, according to the steps 1.1 to 1.4, carrying out sample preparation, spectral data acquisition, characteristic spectral line selection and spectral line combination on rice to be identified to obtain a plurality of groups of rice spectral data;
then, the spectral data of the unknown samples are classified by using the SVM classification models of the rice in different producing areas established in the step 1.5, so that corresponding producing area prediction is obtained.
Example (c):
example 1
1. And (4) preparing a sample.
In this embodiment, 10 rice samples (shown in table 1) from different production places were selected, and the specific names and production places of the samples were as follows: golden dolphin Thailand fragrant rice (Guangdong Buddha), Hubei Temple mountain specialty organic rice (Hubei Zhijiang), Guangxi Bama authentic glutinous rice (Guangxi Bama), Panfu Fengjin rice (Liaoning Panjin), Wuchang rice flower fragrant rice (Wuchang of Black Longjiang), northeast brown rice (Jilin Siping Liao), Bianmin tribute (Anhui Anqing), Xiangchi rice (Hunan peach source), Wannian tribute rice (Shangxi Shandong) and Chongming island rice (Jiangsu Taizhou). The method of the invention is used for classifying the origin of the plants. In order to simplify the analysis process, the rice in different producing areas is respectively numbered, and the label values are as follows in sequence: 1,2,3, 4, 5,6, 7, 8, 9, 10; the corresponding producing areas are called Guangdong (GD), Hubei (HB), Guangxi (GX), Liaoning (LN), Heilongjiang (HLJ), Jilin (JL), Anhui (AH), Hunan (HN), Jiangxi (JX) and Jiangsu (JS) in short by the names of the places. Considering the physical characteristics of a sample, such as the fact that the irregularity of the sample can affect a spectrum signal to a certain extent, the sample is preprocessed before LIBS acquires the signal, rice in different production places is respectively crushed by a crusher to prepare uniform powder samples, and then about 15g of the powder samples are weighed and prepared into sheet samples with uniform thickness by a tablet press at the pressure of 25MPa for LIBS spectrum detection.
2. And collecting spectral data of the rice sample.
The spectral acquisition was carried out in an air environment, the experimental setup is shown in fig. 1. the emitted plasma radiation was collected by the collection head 5 and transmitted to the spectrometer 7(Andor Technology, mecellel 5000, wavelength range 200-975 nm, resolution λ/△ λ 5000 ═ 5000) for spectroscopy, using a Q-switched Nd: YAG pulsed laser 1 (quantum Brilliant B, wavelength 532nm, pulse width 8ns, maximum repetition frequency 10Hz) as excitation source, ICCD9(Andor Technology, itar DH-334T, 1024 × 1024 pixels) for photoelectric conversion of the spectrometer transmitted spectral signals, using an electric displacement platform 11 to control the sample surface to make "bow" movements in the X and Y directions during the spectral acquisition process, in order to prevent air breakdown, the focal point of the laser focusing lens 3 (focal length 15cm) was located 1.27mm below the sample surface, in order to obtain the best spectral intensity and spectral energy ratio, the laser pulsed energy was set to 40mJ, the laser pulsed energy was set to cd 1 and the gate energy set to 10 μ j, and the spectral acquisition was carried out in a cmos process of rice spectrum acquisition, thus the rice spectrum was carried out at 1000 μm 2 μm.
3. And selecting a characteristic spectral line.
As rice contains abundant carbohydrate and mineral elements, 13 characteristic lines of main elements, namely C-N (0, 0)388.34nm, C I247.86 nm, N I746.83nm, O I777.19 nm, H I656.29 nm, C-C (0, 0)516.52nm, Mg II 279.55nm, Mn I403.08nm, Ca I422.67 nm, Si I288.16 nm, Al I394.40nm, Na I588.95nm and K I766.49nm are selected, and specific characteristic lines are shown in Table 2. And (3) dividing the characteristic spectral line intensity value in each group of data by the Ca I422.67 nm spectral line intensity value for normalization to finally obtain 1000 groups of spectral data, wherein each group of spectral data comprises 12 analysis variables.
4. Characteristic line combination
The combination of several characteristic lines (C-N (0, 0)388.34nm + C-N (1, 1)387.14nm + C-N (2, 2)386.19nm + C-N (3, 3)385.47nm + C-N (4, 4)385.09nm, N I746.83nm + N I744.23 nm + N I742.36 nm, O I777.19 nm + O I777.42nm, Mg II 279.55nm + Mg II 280.27nm + Mg I285.21 nm, Mn I403.08nm + Mn I31nm + Mn I403.45 nm, Al I394.40nm + Al I396.15 nm, Na I588.95nm + Na I589.59nm and K I766.49nm + K I769.90 nm) after normalization of the same element intensity, which is the strongest characteristic line for each element and is close to each other, increases the difference between rice of different producing areas, and 1000 sets of spectral data are obtained similarly, each set of spectral data containing 12 analytical variables. As can be seen from FIG. 3, the normalized intensity ratios of the lines of different origins of 7, 8 and 9 have small discrimination at the characteristic line C-N (0, 0)388.34nm, and the differences between the origins are effectively increased by the combination of the characteristic lines C-N (0, 0)388.34nm, C-N (1, 1)387.14nm, C-N (2, 2)386.19nm, C-N (3, 3)385.47nm and C-N (4, 4)385.09 nm. The specific difference values of rice in different producing areas are shown in Table 3, wherein12Refers to the normalized difference between the characteristic line intensities of the producing area 1 and the producing area 2.
And 5, establishing an SVM classification model.
The LIBSVM tool box is used for mathematical modeling under the software environment of Matlab2010b, and the adopted kernel function is a radial basis function, so that the method is suitable for classification of nonlinear data and has high stability. The g parameter in the kernel radial basis function and the penalty factor c parameter of the relaxation variable are main factors influencing the performance of the SVM algorithm, and the core principle of the algorithm is to find out the optimal mapping kernel parameter g and the penalty factor c. In order to train the result to be more convincing and prevent the modeling from over learning, an interactive verification method is adopted to optimize the influence factors. The c and g parameters of the single characteristic spectral line are 222.8609 and 0.125 respectively; the c and g parameters of the combined characteristic line are 588.1336 and 0.047366, respectively. Finally, an SVM training model is used to obtain a training set recognition rate of the single characteristic spectral line of 91.2%, a prediction set recognition rate of 90.8%, a training set recognition rate of a multi-spectral line combination of 94.2%, and a prediction set recognition rate of 94.6%, so that a high recognition rate is obtained, and specific recognition results of different places of production and SVM recognition results combined with spectral lines are shown in a table 4, wherein the recognition rate of Hubei Zhijiang rice is improved from 64% to 88%. Therefore, the accuracy of rice origin identification through laser-induced breakdown spectroscopy can be improved through multi-characteristic spectral line combination.
6. Comparison with other weight value adjustment methods
Comparing the spectral line combination method with the chinese patent document "a method and apparatus for improving the plastic identification accuracy of laser probe" (publication No. CN104730041A, publication date 2015, 6 and 24) discloses a LIBS technique combining spectral line weight adjustment with SVM, by increasing the weight of the nonmetal principal component C, N, H, the result is shown in table 4, and 94.6% of the identification rate of the spectral line combination method for rice is higher than 90.8% of the identification rate of the spectral line weight adjustment.
Example 2
1. And (4) preparing a sample.
In the embodiment, 6 rice samples (as shown in table 5) from the same province and different production places are selected, and the specific product names and the production places of the samples are as follows: seven river sources (the black dragon river is in the state of being seiulic), wuchang rice flower fragrance rice (wuchang black dragon river), baby complementary food (lotus village of Ningan of black dragon river), Jinglin glutinous rice (Qi Ha Er of black dragon river), northeast soundrice (the water village of Ningan of black dragon river) and vegetarian cat Tailai rice (the rice of Qi Ha Er of black dragon river). The method of the invention is used for classifying the origin of the plants. In order to simplify the analysis process, the rice in different producing areas is respectively numbered, and the label values are as follows in sequence: 1,2,3, 4, 5, 6; corresponding producing areas are abbreviated as Seiuzhuiji (SH), Wuchan (WC), lotus flower (LH), ziqihaer 1(QQHE1), rattle (XS) and ziqihaer 2(QQHE 2). And similarly, respectively crushing the rice in different producing areas by using a crusher to prepare uniform powder samples, and then weighing about 15g of the powder samples to prepare sheet samples with uniform thickness by using a tablet press under the pressure of 25MPa for LIBS spectrum detection.
2. And collecting spectral data of the rice sample.
Under the same experimental conditions, spectra were collected from plasmas of 6 kinds of rice, 100 spectra were collected from each sample, and thus 600 spectra were collected from 6 kinds of rice in total.
3. And selecting a characteristic spectral line.
13 characteristic lines of the major elements C-N (0, 0)388.34nm, C I247.86 nm, N I746.83nm, O I777.19 nm, H I656.29 nm, C-C (0, 0)516.52nm, Mg II 279.55nm, Mn I403.08nm, Ca I422.67 nm, Si I288.16 nm, Al I394.40nm, Na I588.95nm and K I766.49nm are also selected. And (3) dividing the characteristic spectral line intensity value in each group of data by the Ca I422.67 nm spectral line intensity value for normalization to finally obtain 600 groups of spectral data, wherein each group of spectral data comprises 12 analysis variables.
4. Characteristic line combination
Similarly, multiple characteristic lines (C-N (0, 0)388.34nm + C-N (1, 1)387.14nm + C-N (2, 2)386.19nm + C-N (3, 3)385.47nm + C-N (4, 4)385.09nm, N I746.83nm + N I744.23 nm + N I742.36 nm, O I777.19 nm + O I777.42nm, Mg II 279.55nm + Mg II 280.27nm + Mg 285.21nm, Mn I403.08nm + Mn I403.31nm + Mn I403.45 nm, Al I394.40nm + Al I396.15 nm, Na I588.95nm + Na I589.59nm and KI766.49nm + KI 769.90nm) are combined after the same element intensity which is the strongest characteristic line of each element and is close to each other is normalized, so that the difference between rice of different producing areas is increased, and 600 sets of spectral data are also obtained, each set of spectral data contains 12 analytical variables.
And 5, establishing an SVM classification model.
The LIBSVM toolkit was used for mathematical modeling in a software environment of Matlab version 2010 b. The c and g parameters of the single characteristic spectral line are 12.1257 and 2.2974 respectively; the c and g parameters of the combined characteristic line are 64 and 0.43528 respectively. Finally, an SVM training model is used to obtain a training set recognition rate of the single characteristic spectral line of 93.67%, a prediction set recognition rate of 88.67%, a training set recognition rate of a multi-spectral line combination of 94.33%, and a prediction set recognition rate of 93%, so that a high recognition rate is obtained, and specific different-place recognition results and SVM recognition results combined with the spectral lines are shown in a table 6. Therefore, the accuracy of rice origin identification through laser-induced breakdown spectroscopy can be improved through multi-characteristic spectral line combination.
6. Comparison with other weight value adjustment methods
The LIBS technical results of comparing the spectral line combination method and spectral line weight adjustment in combination with SVM are shown in table 4, and the recognition rate of the rice by adopting the spectral line combination method is 93% higher than the recognition rate of the spectral line weight adjustment is 92.33%.
Example 3
1. And (4) preparing a sample.
In this embodiment, 10 rice samples (shown in table 7) of 20 different producing areas of different provinces are selected, and the specific product names and producing areas of the samples are as follows: golden dolphin Thailand fragrant rice (Guangdong Buddha), Hubei Guanomishan special organic rice (Hubei Zhijiang), Qihe Yuan (Heilongjiang Sublization), Liangshrimp king fragrant soft rice (Guangdong Dongguan), Tiandi grain man (Liaoning Chaoyang), Guangxi Bama authentic glutinous rice (Guangxi Bama), Panfu Fengjin rice (Liaoning Panjin), five cereals Xiaoxi rice (Guangdong Guangzhou), Wuchang rice fragrance rice (Heilong Wuchang), baby complementary food (Heilongjiang Ningan lotus village), Jinglin glutinous rice (Heilongjiang Qiqi Hara), northeast brown rice (Jilin Siping Liao), Bingmin Yugong (Anhui Anqing), northeast Xiangshui rice (Heilongjiang Ningan Shuichun village) Xiangchi rice (Hunan Taoyuan), Wannian Gongmi rice (Shangxi Shango), Chongming island rice (Jiangsu Taizhou), one Jiangsuang selenium-enriched rice (Jiangxi Jian), Zhuxi Gongmi (Hubei Shiweizhun) and vegetarian cat Tailai rice (Heilongjiang Qiqi Hara). The method of the invention is used for classifying the origin of the plants. In order to simplify the analysis process, the rice in different producing areas is respectively numbered, and the label values are as follows in sequence: 1,2,3, 4, 5,6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20; the corresponding producing areas are abbreviated as Buddha mountain (FS), Zhijiang (ZJ), Suizhihua (SH), Dongguan (DG), Chaoyang (CY), river basin (HC), Panjin (PJ), Guangzhou (GZ), Wuchan (WC), Lihuacun (LHC), Qizihaer (QQHE), Liaoning (SL), Anqing (AQ), Xiangshui (XSC), peach source (TY), Shandong (SR), Taizhou (TZ), Jian (JA), Zhuxi (ZX) and Heilongjiang (HLJ) by the names of places. And similarly, respectively crushing the rice in different producing areas by using a crusher to prepare uniform powder samples, and then weighing about 15g of the powder samples to prepare sheet samples with uniform thickness by using a tablet press under the pressure of 25MPa for LIBS spectrum detection.
2. And collecting spectral data of the rice sample.
Under the same experimental conditions, spectra were collected from plasmas of 20 types of rice, 100 spectra were collected from each sample, and thus 2000 spectra were collected from 20 types of rice.
3. And selecting a characteristic spectral line.
13 characteristic lines of the major elements C-N (0, 0)388.34nm, C I247.86 nm, N I746.83nm, O I777.19 nm, H I656.29 nm, C-C (0, 0)516.52nm, Mg II 279.55nm, Mn I403.08nm, Ca I422.67 nm, Si I288.16 nm, Al I394.40nm, Na I588.95nm and K I766.49nm are also selected. And (3) dividing the characteristic spectral line intensity value in each group of data by the Ca I422.67 nm spectral line intensity value for normalization to finally obtain 2000 groups of spectral data, wherein each group of spectral data comprises 12 analysis variables.
4. Characteristic line combination
Similarly, a plurality of characteristic lines (C-N (0, 0)388.34nm + C-N (1, 1)387.14nm + C-N (2, 2)386.19nm + C-N (3, 3)385.47nm + C-N (4, 4)385.09nm, N I746.83nm + N I744.23 nm + N I742.36 nm, O I777.19 nm + O I777.42nm, Mg II 279.55nm + Mg II 280.27nm + Mg I285.21 nm, Mn I403.08nm + Mn I403.31nm + Mn I403.45 nm, Al I394.40nm + Al I396.15 nm, Na I588.95nm + Na I589.59nm and K I766.49nm + K I769.90 nm) were combined after normalization of the same element intensity, which is the strongest characteristic line of each element and is close to each other, to increase the difference between different rice producing areas, and similarly 2000 sets of spectral data were obtained, each set of spectral data including 12 analytical variables.
And 5, establishing an SVM classification model.
The LIBSVM toolkit was used for mathematical modeling in a software environment of Matlab version 2010 b. The c and g parameters of the single characteristic spectral line are 222.8609 and 0.37893 respectively; the c and g parameters of the combined characteristic line are 388.0234 and 0.071794, respectively. Finally, an SVM training model is used to obtain a training set recognition rate of the single characteristic spectral line, the prediction set recognition rate is 86.7%, a training set recognition rate of the multi-spectral line combination is 90.1%, a prediction set recognition rate is 89.4%, a high recognition rate is obtained, and specific different-place recognition results and SVM recognition result pair ratios combined with the spectral lines are shown in Table 8. Therefore, the accuracy of rice origin identification through laser-induced breakdown spectroscopy can be improved through multi-characteristic spectral line combination. 6. Comparison with other weight value adjustment methods
The LIBS technical results of comparing the spectral line combination method and spectral line weight adjustment in combination with SVM are shown in table 4, and the recognition rate of 89.4% to rice by adopting the spectral line combination method is higher than the recognition rate of 84.8% to spectral line weight adjustment.
The foregoing shows and describes the general principles and features of the present invention, together with the advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are given by way of illustration of the principles of the present invention, and that various changes and modifications may be made without departing from the spirit and scope of the invention as defined by the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Table 1 shows the rice list of different producing areas
TABLE 2 combination of single and multiple characteristic lines
Table 3 shows the specific difference values after normalization of the intensities of the characteristic spectral lines of the rice in different production areas in the combination of the single characteristic spectral line and the multiple characteristic spectral lines
Table 4 shows the SVM classification results of rice from different places of origin
Table 5 shows the rice list of the same province and different producing areas
Table 6 shows SVM classification results of rice of the same identity and different producing areas
Table 7 shows the rice list of different provinces and different producing areas
Table 8 shows SVM classification results for rice of different provinces and different origins
Claims (6)
1. A multispectral combined laser-induced breakdown spectroscopy grain crop production place identification method is characterized by comprising two processes of SVM classification model establishment and production place identification, and the method comprises the following steps:
(1) establishing an SVM classification model:
1.1. sample preparation:
preparing powder samples with the same mass into flaky samples with uniform thickness, wherein the samples are n cereal crops with different producing areas;
1.2. collecting spectral data of a sample:
LIBS spectrum data collection is carried out on N grain crops in different producing areas, m plasma spectrums are collected for each grain crop sample in each producing area, N-N-m plasma spectrums are obtained in total, wherein m is the spectrum magnitude number of each sample collected by the LIBS, and m-1, 2, … and 1000;
1.3. selecting characteristic spectral lines
Analyzing plasma spectrums of n grain crops in different producing areas, setting s-1 to represent the number of indexes to be analyzed, selecting s major elements, taking the strongest characteristic spectral line of each major element as an algorithm analysis index, reading the strength value of each strongest characteristic spectral line, and selecting the strength value of the strongest characteristic spectral line of one major element to perform normalization processing on the strength values of the strongest characteristic spectral lines of other s-1 major elements;
1.4. spectral line combination
Combining the same characteristic spectral lines near the strongest characteristic spectral lines of the selected s principal elements to finally obtain q combined indexes to be analyzed;
1.5.SVM Classification model establishment
Mapping linearly indivisible data into a high-dimensional space in a mode of constructing a kernel mapping function, so as to obtain linear distinction in the high-dimensional space, wherein the kernel mapping function adopts a radial basis function:
selecting partial group spectrums of N groups of cereal crop spectrum data as a training set for establishing a support vector machine model, and using the rest group spectrums as a model test set for testing the identification precision of the established support vector machine model;
(2) origin identification
Firstly, according to the steps from 1.1 to 1.4, carrying out sample preparation, spectral data acquisition, characteristic spectral line selection and spectral line combination on cereal crops to be identified to obtain a plurality of groups of cereal crop spectral data;
then, the spectral data of the unknown samples are classified by using the SVM classification models of the cereal crops in different producing areas established in the step 1.5, so as to obtain corresponding producing area prediction.
2. The method of multispectral combined laser-induced breakdown spectroscopy grain crop location identification as recited in claim 1, wherein the grain crop is rice.
3. A method of multiline combination laser induced breakdown spectroscopy grain crop habitat identification as claimed in claim 1 or 2 wherein said LIBS spectral data acquisition process in step 1.2 is: firstly, selecting grain crops of one production place of n grain crops of different production places to carry out optimization on experimental parameters of acquisition delay time, laser energy and defocusing amount of a spectral signal so as to obtain an LIBS signal with high spectral intensity and signal-to-noise ratio; then, the spectral intensity stability and the signal-to-noise ratio of the major elements in the cereal crops are taken as evaluation indexes, and the optimal experimental conditions are comprehensively determined by optimizing the spectral intensity stability and the maximum signal-to-noise ratio of the major elements; under the best experimental conditions, LIBS spectral data acquisition was performed on n different-producing-area cereal crops.
4. A method of multiline combination laser induced breakdown spectroscopy grain crop habitat identification as claimed in claim 3 wherein said spectral intensity stability is determined by calculating the relative standard deviation RSD; the signal-to-noise ratio is specifically calculated by subtracting the background signal intensity value from the collected spectral intensity value, divided by the noise value, and selecting the signal-to-noise ratio that is the largest.
5. A method of multiline combination laser induced breakdown spectroscopy grain crop production location identification as claimed in claim 1 or 2 wherein step 1.5, let each instance in the training set be (X)i,yi)，i＝1,…,l，Xi,yiRespectively representing the ith sample data and its corresponding tag value, wherein Xi(xi1,xi2,…,xir) E is equal to Rr, Rr represents a training sample data set, and r represents the number of attribute values; y isiE {1,2,3, … }, 1,2,3, … are label values of each set of attribute values;
selecting a radial basis function as a kernel mapping function:
the resulting nonlinear SVM classifier equation is:
constraint conditions are as follows:
wherein,sample data representing a prediction set;a mapping function representing a prediction set sample;representing a distance of two norms,/, representing the number of training sets, αiIs a lagrange multiplier; b is a bias factor in the equation; c is a penalty parameter; g is a mapping kernel parameter.
6. A method of multispectral combined laser-induced breakdown spectroscopy grain crop production location identification as recited in claim 5, wherein c, g are selected in a range of: c 2e，g＝2w，e、w∈{-10,-9,…,9,10}。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610802039.4A CN106404748B (en) | 2016-09-05 | 2016-09-05 | A kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610802039.4A CN106404748B (en) | 2016-09-05 | 2016-09-05 | A kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106404748A CN106404748A (en) | 2017-02-15 |
| CN106404748B true CN106404748B (en) | 2019-03-05 |
Family
ID=57998336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610802039.4A Active CN106404748B (en) | 2016-09-05 | 2016-09-05 | A kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106404748B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10984334B2 (en) * | 2017-05-04 | 2021-04-20 | Viavi Solutions Inc. | Endpoint detection in manufacturing process by near infrared spectroscopy and machine learning techniques |
| CN107525797A (en) * | 2017-07-27 | 2017-12-29 | 上海交通大学 | A kind of LIBS analysis methods of micron dimension powdered rubber trace element |
| CN107677647B (en) * | 2017-09-25 | 2021-05-25 | 重庆邮电大学 | Method for identifying origin of traditional Chinese medicinal materials based on principal component analysis and BP neural network |
| JP7465807B2 (en) * | 2017-12-28 | 2024-04-11 | スリープ ナンバー コーポレイション | Bed with presence sensing features |
| WO2019178850A1 (en) * | 2018-03-23 | 2019-09-26 | Oppo广东移动通信有限公司 | Wireless communication method, user equipment and network device |
| WO2019222964A1 (en) * | 2018-05-24 | 2019-11-28 | 深圳达闼科技控股有限公司 | Method for determining detection equipment, detection device and readable storage medium |
| CN109142251B (en) * | 2018-09-17 | 2020-11-03 | 平顶山学院 | LIBS quantitative analysis method of random forest auxiliary artificial neural network |
| CN109916991A (en) * | 2019-04-09 | 2019-06-21 | 新疆大学 | A method of based on metallic element combination PLS-DA Model checking hop varieties and the place of production |
| CN111401444B (en) * | 2020-03-16 | 2023-11-03 | 深圳海关食品检验检疫技术中心 | Method and device for predicting red wine origin, computer equipment and storage medium |
| CN112051256B (en) * | 2020-07-22 | 2023-01-24 | 中国地质大学（武汉） | CNN model-based LIBS (laser induced breakdown spectroscopy) measurement method and system for content of element to be measured |
| CN111965167A (en) * | 2020-08-20 | 2020-11-20 | 天津大学 | Method and device for predicting element composition and calorific value of solid waste |
| CN112782151B (en) * | 2021-02-22 | 2023-01-13 | 湖北工程学院 | Data processing method for improving classification accuracy of laser-induced breakdown spectroscopy |
| CN114579635B (en) * | 2022-03-04 | 2022-11-04 | 北京三月雨文化传播有限责任公司 | Big data information analysis processing system based on cloud computing |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488874A (en) * | 2013-09-01 | 2014-01-01 | 西北大学 | Improved support vector machine-LIBS (laser-induced breakdown spectroscopy) combined sorting method for steel materials |
| CN104697965A (en) * | 2015-03-10 | 2015-06-10 | 西北大学 | Method for recognizing slag variety by combining with laser-induced breakdown spectroscopy based on least squares support vector machine |
| CN104730041A (en) * | 2013-12-20 | 2015-06-24 | 武汉新瑞达激光工程有限责任公司 | Method and apparatus for improving plastic identification precision of laser probe |
| CN104964950A (en) * | 2015-06-10 | 2015-10-07 | 长江大学 | Multi-element wave peak-based laser-induced breakdown spectroscopy rock fragment type identification method |
| CN105181678A (en) * | 2015-09-07 | 2015-12-23 | 长江大学 | Identification method of rice varieties based on laser-induced breakdown spectroscopy (LIBS) |
| CN105806827A (en) * | 2016-03-11 | 2016-07-27 | 华中科技大学 | Method for identifying plastics by virtue of laser probe based on non-metallic element |
-
2016
- 2016-09-05 CN CN201610802039.4A patent/CN106404748B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488874A (en) * | 2013-09-01 | 2014-01-01 | 西北大学 | Improved support vector machine-LIBS (laser-induced breakdown spectroscopy) combined sorting method for steel materials |
| CN104730041A (en) * | 2013-12-20 | 2015-06-24 | 武汉新瑞达激光工程有限责任公司 | Method and apparatus for improving plastic identification precision of laser probe |
| CN104697965A (en) * | 2015-03-10 | 2015-06-10 | 西北大学 | Method for recognizing slag variety by combining with laser-induced breakdown spectroscopy based on least squares support vector machine |
| CN104964950A (en) * | 2015-06-10 | 2015-10-07 | 长江大学 | Multi-element wave peak-based laser-induced breakdown spectroscopy rock fragment type identification method |
| CN105181678A (en) * | 2015-09-07 | 2015-12-23 | 长江大学 | Identification method of rice varieties based on laser-induced breakdown spectroscopy (LIBS) |
| CN105806827A (en) * | 2016-03-11 | 2016-07-27 | 华中科技大学 | Method for identifying plastics by virtue of laser probe based on non-metallic element |
Non-Patent Citations (8)
| Title |
|---|
| Classification of steel materials by laser-induced breakdown spectroscopy coupled with support vector machines;Long Liang et al.;《Applied Optics》;20140201;第53卷(第4期);第544-552页 |
| Improving data stability and prediction accuracy in laser-induced breakdown spectroscopy by utilizing a combined atomic and ionic line algorithm;Zongyu Hou et al.;《Journal of Analytical Atomic Spectrometry》;20130630;第29卷(第1期);第107-113页 |
| Incorporation of Support Vector Machines in the LIBS Toolbox for Sensitive and Robust Classification Amidst Unexpected Sample and System Variability;Narahara Chari Dingari et al.;《Analytical Chemistry》;20120320;第84卷(第6期);第2686-2694页 |
| Tianlong Zhang et al..Quantitative and classi fication analysis of slag samples by laser induced breakdown spectroscopy (LIBS) coupled with support vector machine (SVM) and partial least square (PLS) methods.《Journal of Analytical Atomic Spectrometry》.2015,第30卷(第2期),第368-374页. |
| 农产品产地溯源技术研究进展;张晓焱 等;《食品科学》;20101231;第31卷(第3期);第271-278页 |
| 双谱线内标对激光诱导击穿光谱稳定性的改善;刘莉;《激光技术》;20150131;第39卷(第1期);第90-95页 |
| 基于近红外光谱和模式识别技术鉴别大米产地的研究;夏立娅 等;《光谱学与光谱分析》;20130131;第33卷(第1期);第102-105页 |
| 支持向量机算法在激光诱导击穿光谱技术塑料识别中的应用研究;于洋 等;《物理学报》;20131231;第62卷(第21期);第215201-1-215201-7页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106404748A (en) | 2017-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106404748B (en) | A kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition method | |
| CN106770194B (en) | Cereal crops place of production discrimination method based on wavelet transformation laser induced breakdown spectroscopy | |
| Weng et al. | Hyperspectral imaging for accurate determination of rice variety using a deep learning network with multi-feature fusion | |
| CN101881726B (en) | Nondestructive detection method for comprehensive character living bodies of plant seedlings | |
| Feng et al. | Preliminary study on classification of rice and detection of paraffin in the adulterated samples by Raman spectroscopy combined with multivariate analysis | |
| CN103411906B (en) | The near infrared spectrum qualitative identification method of pearl powder and oyster shell whiting | |
| CN107515203A (en) | The research of near infrared technology quantitative analysis rice single grain amylose content | |
| CN104142311B (en) | A kind of method of near-infrared spectrum technique prediction torch pine pine resin yield | |
| CN105067531A (en) | Mango quality nondestructive detection method and mango quality nondestructive detection apparatus | |
| Cui et al. | Identification of maize seed varieties based on near infrared reflectance spectroscopy and chemometrics | |
| CN104155264A (en) | Method for predicting content of turpentine in loblolly pine gum by using near infrared spectroscopy | |
| CN104020128A (en) | Method for rapidly identifying propolis source | |
| CN106383088A (en) | A seed purity rapid nondestructive testing method based on a multispectral imaging technique | |
| CN109520965A (en) | A method of lysine content is detected based near infrared spectrum characteristic extractive technique | |
| CN106940292A (en) | Bar denier wood raw material quick nondestructive discrimination method of damaging by worms based on multi-optical spectrum imaging technology | |
| CN114112983B (en) | Tibetan medicine all-edge leaf green chorine artemisia origin discrimination method based on Python data fusion | |
| Aredo et al. | Predicting of the quality attributes of orange fruit using hyperspectral images | |
| Guifang et al. | Identification of varieties of natural textile fiber based on Vis/NIR spectroscopy technology | |
| CN115905881B (en) | Yellow pearl classification method and device, electronic equipment and storage medium | |
| CN117949429A (en) | Almond origin identification method based on Raman spectrum and multi-mode mixed model | |
| Yang et al. | Collaborative estimation of heavy metal stress in wheat seedlings based on LIBS-Raman spectroscopy coupled with machine learning | |
| Zhou et al. | Germination prediction of sugar beet seeds based on HSI and SVM-RBF | |
| Hashim et al. | Classification of rubber tree leaf disease using spectrometer | |
| CN114878549A (en) | Corn aflatoxin detection method based on Raman spectrum | |
| CN109406421B (en) | Method for predicting ferulic acid content in wolfberry fruit based on hyperspectral imaging technology |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Li Xiangyou Inventor after: Yang Ping Inventor after: Guo Lianbo Inventor after: Zhu Yining Inventor after: Zeng Xiaoyan Inventor after: Li Jiaming Inventor after: Yang Xinyan Inventor after: Tang Yun Inventor before: Li Xiangyou Inventor before: Yang Ping Inventor before: Guo Lianbo Inventor before: Zhu Yining Inventor before: Zeng Xiaoyan Inventor before: Lu Yongfeng Inventor before: Li Jiaming Inventor before: Yang Xinyan Inventor before: Tang Yun |
|
| CB03 | Change of inventor or designer information |