CN111380844A - Method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology - Google Patents
Method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology Download PDFInfo
- Publication number
- CN111380844A CN111380844A CN202010031988.3A CN202010031988A CN111380844A CN 111380844 A CN111380844 A CN 111380844A CN 202010031988 A CN202010031988 A CN 202010031988A CN 111380844 A CN111380844 A CN 111380844A
- Authority
- CN
- China
- Prior art keywords
- matrix
- edible oil
- spectral clustering
- oil
- spectrum data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
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
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N21/6402—Atomic fluorescence; Laser induced fluorescence
-
- 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
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
-
- 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
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/64—Fluorescence; Phosphorescence
- G01N2021/6417—Spectrofluorimetric devices
Abstract
The invention relates to a method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology, which comprises the following working steps: 101, collecting samples of different types of edible oil and making labels; 102, acquiring original fluorescence spectrum data of different types of edible oil by using laser-induced fluorescence equipment; 103, denoising original fluorescence spectrum data of different types of edible oil by adopting a sliding smoothing method; 104 using spectral clustering algorithm for modeling; 105, optimizing a parameter sigma of an exponential kernel function in the spectral clustering model by using an MFO algorithm; and 106, verifying the result. The method adopts the LIF technology combined with spectral clustering and uses the MFO algorithm to optimize the parameter sigma of the spectral clustering to identify the type of the edible oil, has high identification rate, strong generalization capability and practical application value, and is suitable for nondestructive identification of the edible oil.
Description
Technical Field
The invention relates to the field of national food safety, in particular to a method for identifying edible oil by combining spectral clustering with a Laser Induced Fluorescence (LIF) technology.
Background
Edible oil is one of the important sources of fatty acid of human body, and different edible oil types provide different nutrient components for human body. In recent years, the quality of edible oil is in endless, so that many lawless persons can use low-price rapeseed oil and the like to supplement high-price sunflower seed oil and the like to gain benefits, and even more, illegal cooking oil is used to supplement edible oil, so that not only is the legal regulation violated, but also the illegal cooking oil seriously harms the physical health of people.
The traditional edible oil identification methods comprise a gas chromatography method, a liquid chromatography method, an electronic nose method and the like, the identification process is long in time consumption, the equipment is heavy and high in price, the popularization of edible oil identification is not facilitated, and the strength of hitting molecules is limited.
The Laser Induced Fluorescence (LIF) technology is a newly developed technology, and the technology for collecting the spectral data of the edible oil has the advantages of high speed, convenient operation and the like. Spectral clustering is an algorithm based on graph theory, labels of edible oil fluorescence data are not needed, and the identification accuracy is high. The spectral clustering and the laser-induced fluorescence technology have important research value in the identification of the edible oil.
Disclosure of Invention
The invention aims to provide a method for identifying edible oil by combining spectral clustering with a Laser Induced Fluorescence (LIF) technology, and aims to realize nondestructive identification of the edible oil.
The invention adopts the following technical scheme for realizing the purpose:
a method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology comprises the following working steps:
101 samples of different kinds of edible oils were collected: collecting five edible oil samples of soybean oil, sesame oil, corn oil, rapeseed oil and sunflower seed oil, and making labels;
102 acquiring raw fluorescence spectrum data of a sample: acquiring fluorescence spectrum data of the five samples by using laser-induced fluorescence experimental equipment to obtain original fluorescence spectrum data of the edible oil sample;
103, preprocessing raw fluorescence spectrum data: the original fluorescence spectra of five collected edible oil samples contain certain noise, and the existence of the noise can influence the identification accuracy, so that the original fluorescence spectrum data of the samples needs to be subjected to denoising treatment by adopting a Moving-Average method (Moving-Average);
104 spectral clustering modeling: establishing a model by using a spectral clustering algorithm to identify the processed oil sample fluorescence spectrum data;
optimization of 105MFO parameters: the selection of the parameters of the spectral clustering model has great influence on the identification effect of the edible oil sample, and an MFO optimization algorithm is required to optimize the parameter sigma of the exponential kernel function in the spectral clustering model;
and 106, verifying the result.
In the working step 102, the laser-induced fluorescence testing apparatus includes: the laser uses BL-450/1-1500mW blue light laser, the spectrometer uses ocean optics USB micro optical fiber spectrometer, the fluorescence spectrum wave band for collecting five kinds of edible oil is 340-1021 nm.
And when the fluorescence spectrum data of the oil sample is removed by the sliding average method in the working step 103, taking a value of the window width as 3.
In the working step 104, a spectral clustering algorithm is used to establish a model to identify the processed oil sample fluorescence spectrum data, and the specific optimization steps are as follows:
401 construct a similarity matrix W: construction of an element W in a similarity matrix W, W using an exponential kernel functionijThe calculation formula of (a) is as follows:
x in the formulaiAnd XjThe data samples are the ith data sample and the j data sample, and sigma is an exponential kernel function parameter;
402 constructs a pair angle matrix D: element D in DiThe calculation formula of (a) is as follows:
403 constructs a laplacian matrix L: the formula is as follows:
L=D-W
404 construct a feature matrix F: the Laplace matrix L is normalized to obtain L', and the first 5 minimum eigenvectors are calculated to form an eigenvector matrix F according to rows, and the normalization formula is as follows:
L′=D-0.5LD-0.5
405 accuracy of the output model: and taking the characteristic matrix F as the input of k-means, clustering and outputting the clustering accuracy.
In the working step 105, an MFO algorithm is used to optimize a parameter σ of an exponential kernel function in a spectral clustering model, and the specific optimization steps are as follows:
501: initializing parameters of the MFO: selection of the range of the kernel parameter σ [ a ]-q,aq]Training with the maximum iteration number set as m and the maximum moth number and flame number set as n;
502: and (3) generating a moth matrix M and a moth fitness matrix OM: generating a moth matrix M according to the range of the kernel function parameter sigma and the number of moths, taking the spectral clustering model as a fitness function of the moths, and forming a fitness matrix OM of the moths according to the accuracy of the model;
503: generating a flame matrix F and a flame fitness matrix OF: performing ascending sequencing on the moth matrix M and the fitness matrix OM OF the moths to obtain a corresponding flame matrix F and a corresponding fitness matrix OF OF the flames;
504: and (4) updating M: continuously updating the moth matrix M, wherein the updating formula is as follows:
S(Mi,Fj)=Di·ebt·cos(2πt)+Fj
in the formula DiRefers to the distance between the ith moth and the jth flame, FjRefers to the jth flame, b is the constant of the logarithmic spiral, t is a constant within [ -2,1 [ ]]A random number in between;
505: and (3) reducing the number of flames: reducing the number of flames, the update formula is as follows:
wherein n refers to the maximum number of flames, m refers to the maximum iteration number, and l refers to the current iteration number;
506: whether the number of iterations is reached: if the maximum number of iterations is reached, the MFO optimization process is ended, otherwise execution continues from step 402;
507: optimal value of output parameter σ: and outputting the optimal value of the exponential kernel function parameter sigma after the maximum iteration times are reached.
Has the advantages that:
compared with the prior art, the invention has the beneficial effects that: the laser-induced fluorescence technology has the advantages of high speed, convenience in operation and the like. The performance of the spectral clustering algorithm has a great relationship with the selection of the parameters, and the MFO algorithm is used for optimizing the parameters of the spectral clustering algorithm, so that the influence caused by unreasonable parameter selection can be avoided. The identification of the edible oil by combining spectral clustering with a laser-induced fluorescence technology has the advantages of short time consumption, high identification precision and realization of nondestructive identification of the edible oil compared with the traditional gas chromatography and liquid chromatography. The model has strong generalization capability and practical application value, and is very suitable for the nondestructive identification of the edible oil.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of an experimental apparatus for collecting spectra of an edible oil sample according to the present invention;
FIG. 3 is a flow chart of an algorithm for spectral clustering modeling in accordance with the present invention;
FIG. 4 is a flow chart of an algorithm for optimizing spectral clustering parameters using MFO in accordance with the present invention;
FIG. 5 is a graph of model identification accuracy.
Detailed Description
The invention is further illustrated by the following specific examples.
The invention has developed a spectral clustering and combined the Laser Induced Fluorescence (LIF) technology and discerns the method of the edible oil, gather five kinds of edible oil samples of soybean oil, sesame oil, corn oil, rapeseed oil and sunflower seed oil at first; secondly, collecting fluorescence spectrum data of the five edible oils by using laser-induced fluorescence equipment; then, denoising the fluorescence spectrum of the collected edible oil sample by using a sliding smoothing method, so as to reduce the influence of noise on identification; then establishing a model by using a spectral clustering algorithm; and meanwhile, optimizing the parameters of the spectral clustering by using an MFO algorithm, and finally obtaining an edible oil identification model with better performance.
The invention provides an edible oil identification method on the premise of organically combining spectral clustering and laser-induced fluorescence technology, which comprises the following specific steps as shown in figure 2:
101 samples of different kinds of edible oils were collected: collecting five edible oil samples of soybean oil, sesame oil, corn oil, rapeseed oil and sunflower seed oil, and making labels;
102 acquiring raw fluorescence spectrum data of a sample: acquiring fluorescence spectrum data of five samples by using laser-induced fluorescence experimental equipment shown in figure 2 to obtain original fluorescence spectrum data of the samples;
103, preprocessing raw fluorescence spectrum data: the original fluorescence spectra of five collected edible oil samples contain certain noise, and the existence of the noise can influence the accuracy rate of identification, so that the original fluorescence spectrum data of the samples need to be subjected to denoising treatment by adopting a Moving-Average method (Moving-Average);
104 spectral clustering modeling: the spectral clustering algorithm is used for establishing a model to identify the processed oil sample fluorescence spectrum data, and the specific steps are shown in figure 3:
401 construct a similarity matrix W: construction of an element W in a similarity matrix W, W using an exponential kernel functionijThe calculation formula of (a) is as follows:
x in the formulaiAnd XjThe data samples are the ith data sample and the j data sample, and sigma is an exponential kernel function parameter;
402 constructs a pair angle matrix D: element D in DiThe calculation formula of (a) is as follows:
403 constructs a laplacian matrix L: the formula is as follows:
L=D-W
404 construct a feature matrix F: the Laplace matrix L is normalized to obtain L', and the first 5 minimum eigenvectors are calculated to form an eigenvector matrix F according to rows, and the normalization formula is as follows:
L′=D-0.5LD-0.5
405 accuracy of the output model: and taking the characteristic matrix F as the input of k-means, clustering and outputting the clustering accuracy.
Optimization of 105MFO parameters: the spectral clustering algorithm is used for establishing a model to identify the processed oil sample fluorescence spectrum data, the selection of spectral clustering parameters has great influence on the identification effect, an MFO optimization algorithm is required to optimize the parameter sigma of the exponential kernel function in the spectral clustering, and the specific steps are shown in FIG. 4:
501 initialize the parameters of the MFO: selection of the range of the kernel parameter σ [ a ]-q,aq]Training with the maximum iteration number set as m and the maximum moth number and flame number set as n;
502, generating a moth matrix M and a moth fitness matrix OM: generating a moth matrix M according to the range of the kernel function parameter sigma and the number of moths, taking the spectral clustering model as a fitness function of the moths, and forming a fitness matrix OM of the moths according to the accuracy of the model;
503 generate a flame matrix F and a flame fitness matrix OF: performing ascending sequencing on the moth matrix M and the fitness matrix OM OF the moths to obtain a corresponding flame matrix F and a corresponding fitness matrix OF OF the flames;
504 updates M: continuously updating the moth matrix M, wherein the updating formula is as follows:
S(Mi,Fj)=Di·ebt·cos(2πt)+Fj
in the formula DiRefers to the distance between the ith moth and the jth flame, FjRefers to the jth flame, b is the constant of the logarithmic spiral, t is a constant within [ -2,1 [ ]]A random number in between;
505 reduction of the number of flames: reducing the number of flames, the update formula is as follows:
wherein n refers to the maximum number of flames, m refers to the maximum iteration number, and l refers to the current iteration number;
506 if the number of iterations is reached: if the maximum number of iterations is reached, the MFO optimization process is ended, otherwise execution continues from step 402;
507 optimal value of the parameter σ: and outputting the optimal value of the exponential kernel function parameter sigma after the maximum iteration times are reached.
106, result verification: in order to verify the effectiveness of the method, 150 groups of spectral data are respectively collected for 5 edible oil samples, 750 groups of fluorescence data are used as the input of a spectral clustering and laser-induced fluorescence technology model, the value of an exponential kernel function parameter sigma after the MFO parameter is optimized is 38.3074, and the identification accuracy is 100% as shown in FIG. 5.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (5)
1. A method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology is characterized by comprising the following steps: the method comprises the following working steps:
101 samples of different kinds of edible oils were collected: collecting five edible oil samples of soybean oil, sesame oil, corn oil, rapeseed oil and sunflower seed oil, and making labels;
102 acquiring raw fluorescence spectrum data of a sample: acquiring fluorescence spectrum data of the five samples by using laser-induced fluorescence experimental equipment to obtain original fluorescence spectrum data of the edible oil sample;
103, preprocessing raw fluorescence spectrum data: the original fluorescence spectra of five collected edible oil samples contain certain noise, and the existence of the noise can influence the identification accuracy, so that the original fluorescence spectrum data of the samples needs to be subjected to denoising treatment by adopting a Moving-Average method (Moving-Average);
104 spectral clustering modeling: establishing a model by using a spectral clustering algorithm to identify the processed oil sample fluorescence spectrum data;
optimization of 105MFO parameters: the selection of the parameters of the spectral clustering model has great influence on the identification effect of the edible oil sample, and an MFO optimization algorithm is required to optimize the parameter sigma of the exponential kernel function in the spectral clustering model;
and 106, verifying the result.
2. The method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology according to claim 1, wherein the method comprises the following steps: in operation 102, the laser induced fluorescence testing apparatus comprises: the laser uses BL-450/1-1500mW blue laser, the spectrometer uses ocean optics USB micro optical fiber spectrometer, the fluorescence spectrum wave band for collecting five kinds of edible oil is 340-1021 nm.
3. The method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology according to claim 1, wherein the method comprises the following steps: in the working step 103, when the fluorescence spectrum data of the oil sample is removed by the sliding average method, the value of the window width is 3.
4. The method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology according to claim 1, wherein the method comprises the following steps: in the working step 104, a spectral clustering algorithm is used to establish a model to identify the processed oil sample fluorescence spectrum data, and the specific optimization steps are as follows:
401 construct a similarity matrix W: construction of an element W in a similarity matrix W, W using an exponential kernel functionijThe calculation formula of (a) is as follows:
x in the formulaiAnd XjThe data samples are the ith data sample and the j data sample, and sigma is an exponential kernel function parameter;
402 constructs a pair angle matrix D: element D in DiThe calculation formula of (a) is as follows:
403 constructs a laplacian matrix L: the formula is as follows:
L=D-W
404 construct a feature matrix F: the laplace matrix L is normalized to obtain L', and the normalization formula for the feature matrix F composed of the first 5 smallest eigenvectors by row is found as follows:
L′=D-0.5LD-0.5
405 accuracy of the output model: and taking the characteristic matrix F as the input of k-means, clustering and outputting the clustering accuracy.
5. The method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology according to claim 1, wherein the method comprises the following steps: in the working step 105, an MFO algorithm is used to optimize a parameter σ of an exponential kernel function in a spectral clustering model, and the specific optimization steps are as follows:
501: initializing parameters of the MFO: selection of the range of the kernel parameter σ [ a ]-q,aq]Training with the maximum iteration number set as m and the maximum moth number and flame number set as n;
502: and (3) generating a moth matrix M and a moth fitness matrix OM: generating a moth matrix M according to the range of the kernel function parameter sigma and the number of moths, taking the spectral clustering model as a fitness function of the moths, and forming a fitness matrix OM of the moths according to the accuracy of the model;
503: generating a flame matrix F and a flame fitness matrix OF: performing ascending sequencing on the moth matrix M and the fitness matrix OM OF the moths to obtain a corresponding flame matrix F and a corresponding fitness matrix OF OF the flames;
504: and (4) updating M: continuously updating the moth matrix M, wherein the updating formula is as follows:
S(Mi,Fj)=Di·ebt·cos(2πt)+Fj
in the formula DiRefers to the distance between the ith moth and the jth flame, FjRefers to the jth flame, b is the constant of the logarithmic spiral, t is a constant of [ -2,1 ]]A random number in between;
505: and (3) reducing the number of flames: reducing the number of flames, the update formula is as follows:
wherein n refers to the maximum number of flames, m refers to the maximum iteration number, and l refers to the current iteration number;
506: whether the number of iterations is reached: if the maximum number of iterations is reached, the MFO optimization process is ended, otherwise execution continues from step 402;
507: optimal value of output parameter σ: and outputting the optimal value of the exponential kernel function parameter sigma after the maximum iteration times are reached.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010031988.3A CN111380844A (en) | 2020-01-13 | 2020-01-13 | Method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010031988.3A CN111380844A (en) | 2020-01-13 | 2020-01-13 | Method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111380844A true CN111380844A (en) | 2020-07-07 |
Family
ID=71218586
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010031988.3A Pending CN111380844A (en) | 2020-01-13 | 2020-01-13 | Method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111380844A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009080049A1 (en) * | 2007-12-21 | 2009-07-02 | Dma Sorption Aps | Monitoring oil condition and/or quality, on-line or at-line, based on chemometric data analysis of flourescence and/or near infrared spectra |
CN106383103A (en) * | 2016-09-18 | 2017-02-08 | 中国科学院海洋研究所 | Time-resolved oil fluorescence identification method based on main component analysis |
CN107462560A (en) * | 2017-09-08 | 2017-12-12 | 周茗皓 | LIF combinations LIBS edible oil quality fast analyser and method |
CN108827925A (en) * | 2018-08-01 | 2018-11-16 | 滁州学院 | Edible vegetable oil true and false rapid detection method and detection device based on optical fiber type fluorescence spectroscopy technique |
CN109598296A (en) * | 2018-11-26 | 2019-04-09 | 长安大学 | One kind is based on a flying moth darts into the fire the K mean cluster method of improvement |
CN110398472A (en) * | 2019-06-28 | 2019-11-01 | 天津大学 | A kind of vegetable oil identification classification method based on two-dimensional correlation spectra quantization |
-
2020
- 2020-01-13 CN CN202010031988.3A patent/CN111380844A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009080049A1 (en) * | 2007-12-21 | 2009-07-02 | Dma Sorption Aps | Monitoring oil condition and/or quality, on-line or at-line, based on chemometric data analysis of flourescence and/or near infrared spectra |
CN106383103A (en) * | 2016-09-18 | 2017-02-08 | 中国科学院海洋研究所 | Time-resolved oil fluorescence identification method based on main component analysis |
CN107462560A (en) * | 2017-09-08 | 2017-12-12 | 周茗皓 | LIF combinations LIBS edible oil quality fast analyser and method |
CN108827925A (en) * | 2018-08-01 | 2018-11-16 | 滁州学院 | Edible vegetable oil true and false rapid detection method and detection device based on optical fiber type fluorescence spectroscopy technique |
CN109598296A (en) * | 2018-11-26 | 2019-04-09 | 长安大学 | One kind is based on a flying moth darts into the fire the K mean cluster method of improvement |
CN110398472A (en) * | 2019-06-28 | 2019-11-01 | 天津大学 | A kind of vegetable oil identification classification method based on two-dimensional correlation spectra quantization |
Non-Patent Citations (2)
Title |
---|
王翔等: "基于激光诱导荧光光谱的塑料分类鉴别", 《光电子▪激光》 * |
韩仲志等: "利用油品紫外荧光特性的多光谱成像检测", 《发光学报》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110287334B (en) | Method for constructing knowledge graph in school domain based on entity identification and attribute extraction model | |
CN111126386B (en) | Sequence domain adaptation method based on countermeasure learning in scene text recognition | |
CN109299751B (en) | EMD data enhancement-based SSVEP electroencephalogram classification method of convolutional neural model | |
CN110807760B (en) | Tobacco leaf grading method and system | |
CN108362659B (en) | Edible oil type rapid identification method based on multi-source spectrum parallel fusion | |
He et al. | Rethinking Image Aesthetics Assessment: Models, Datasets and Benchmarks. | |
CN109344759A (en) | A kind of relatives' recognition methods based on angle loss neural network | |
CN106845528A (en) | A kind of image classification algorithms based on K means Yu deep learning | |
CN108509601B (en) | Food flavor assessment method based on big data analysis | |
Li et al. | Identification of soybean varieties based on hyperspectral imaging technology and one‐dimensional convolutional neural network | |
CN109359550A (en) | Language of the Manchus document seal Abstraction and minimizing technology based on depth learning technology | |
CN110263174A (en) | - subject categories the analysis method based on focus | |
CN113378680A (en) | Intelligent database building method for Raman spectrum data | |
CN111613267A (en) | CRISPR/Cas9 off-target prediction method based on attention mechanism | |
CN113033066A (en) | Method for establishing near infrared spectrum identification model of sargassum fusiforme production area, strain and cultivation mode and identification method | |
CN109308498B (en) | Identification method of laser-induced fluorescent vegetable oil-doped millet soup | |
CN109142265B (en) | Method for rapidly identifying type of edible oil by near infrared spectrum wavelet fusion | |
CN109001181B (en) | Method for rapidly identifying type of edible oil by combining Raman spectrum typical correlation analysis | |
CN114399661A (en) | Instance awareness backbone network training method | |
CN111380844A (en) | Method for identifying edible oil by combining spectral clustering with Laser Induced Fluorescence (LIF) technology | |
CN112712857A (en) | Method for generating biological Raman spectrum data based on WGAN (WGAN) antagonistic generation network | |
Hong et al. | Classification of Oil Palm fruit Ripeness Using Artificial Neural Network | |
CN105447513B (en) | A kind of file ink data realizes the method and system of automatic contrast | |
CN113378673B (en) | Semi-supervised electroencephalogram signal classification method based on consistency regularization | |
CN115375502B (en) | Overlapped community intelligent mining method and system based on double-scale graph wavelet neural network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200707 |