CN111879752A - Ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning - Google Patents
Ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning Download PDFInfo
- Publication number
- CN111879752A CN111879752A CN202010777408.5A CN202010777408A CN111879752A CN 111879752 A CN111879752 A CN 111879752A CN 202010777408 A CN202010777408 A CN 202010777408A CN 111879752 A CN111879752 A CN 111879752A
- Authority
- CN
- China
- Prior art keywords
- sample
- machine learning
- raman
- data
- probe
- 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
- 239000000523 sample Substances 0.000 title claims abstract description 179
- 238000010801 machine learning Methods 0.000 title claims abstract description 61
- 238000001514 detection method Methods 0.000 title claims abstract description 45
- 238000001237 Raman spectrum Methods 0.000 title claims abstract description 28
- 238000005070 sampling Methods 0.000 title claims abstract description 25
- 238000001069 Raman spectroscopy Methods 0.000 claims abstract description 68
- 230000003287 optical effect Effects 0.000 claims abstract description 53
- 238000004458 analytical method Methods 0.000 claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 32
- 239000013307 optical fiber Substances 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims description 31
- 238000005057 refrigeration Methods 0.000 claims description 17
- 238000002372 labelling Methods 0.000 claims description 16
- 238000000034 method Methods 0.000 claims description 15
- 238000007405 data analysis Methods 0.000 claims description 11
- 230000003044 adaptive effect Effects 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000000513 principal component analysis Methods 0.000 claims description 9
- 238000012706 support-vector machine Methods 0.000 claims description 9
- 238000013524 data verification Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 239000004065 semiconductor Substances 0.000 claims description 6
- 230000000295 complement effect Effects 0.000 claims description 5
- 238000005139 ultra-violet Raman spectroscopy Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 238000001228 spectrum Methods 0.000 description 8
- 230000008030 elimination Effects 0.000 description 6
- 238000003379 elimination reaction Methods 0.000 description 6
- 239000000126 substance Substances 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000005284 excitation Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000002360 preparation method Methods 0.000 description 3
- 238000003672 processing method Methods 0.000 description 2
- 230000005457 Black-body radiation Effects 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 150000004945 aromatic hydrocarbons Chemical group 0.000 description 1
- 150000004951 benzene Chemical class 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000005622 photoelectricity Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
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
- G01N21/65—Raman scattering
-
- 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
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
Landscapes
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention relates to an ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning, which comprises a first optical probe, a second optical probe, a third optical probe and a fourth optical probe, wherein the first optical probe is in contact with the surface of a sample through a laser beam and collects Raman signals of the sample to realize refined detection and analysis and store sample information; the second optical probe is contacted with the surface of the sample through the laser beam and collects a Raman signal of the sample, so that practical and rapid detection and analysis are realized; and the spectrometer is connected with the first optical probe and the second optical probe through optical fibers and is used for receiving the Raman signals of the sample, processing the signals and displaying the signals. By adopting the ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning, disclosed by the invention, the original Raman data acquired by the Raman spectrometer can be respectively subjected to fine analysis and practicability analysis, and automatic and rapid denoising and effective signal identification and classification are realized by combining a self-adaptive machine learning algorithm.
Description
Technical Field
The invention relates to the field of optics, in particular to the technical field of optical detection, and specifically relates to an ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning.
Background
Raman spectroscopy was a technique that was raman-extracted by indian scientists in the last 20 th century. The emergence of lasers in the sixties of the last century brought development space for the application of Raman spectroscopy. The Raman spectrum analysis technology has the advantages of abundant material characteristic information, nondestructive inspection, no need of sample preparation and the like, is widely applied at present as an effective detection analysis method, and is an important component of modern material characteristic analysis technology. The early Raman spectrometer is large in size and inconvenient to install, is mainly applied to laboratories, and along with continuous progress of photoelectricity, embedded software and hardware, a machine learning algorithm and a Raman spectrum analysis technology, the handheld portable Raman spectrometer which is small in size, light in weight and convenient to move has the advantages of being fast and accurate in detection and the like, is rapidly developed, and market demands are continuously increased. Various models of portable raman spectrometers are successively introduced by various known spectrometer manufacturers in the world.
Ultraviolet raman has many advantages, one is to avoid fluorescence interference because the laser excitation wavelength is in the short wavelength region of the ultraviolet band, and because the raman spectrum signal is weak, the excitation light of visible wavelength can easily generate serious fluorescence interference of the object to be measured or impurities. And secondly, the Raman signal has high sensitivity, and theoretical calculation shows that the intensity of the Raman signal is in direct proportion to the fourth power of the frequency, and the frequency of ultraviolet light is much higher than that of visible light and infrared light, so that additional signal enhancement can be brought. Thirdly, the selectivity is very good. The electron energy level differs for different samples or different structures in a sample, and therefore the excitation light that can cause them to resonate differs. The structure of dangerous chemicals generally contains an aromatic hydrocarbon structure of a benzene ring, and can be excited in an electron absorption band, namely, resonance excitation, so that a larger dipole moment can be generated, and the Raman signal intensity is larger.
The original raman spectrum contains noise such as fluorescence, various interference sources, and the like, and greatly influences the measurement. The Raman spectrum data processing technology is used for eliminating fluorescence and noise of a measured spectrum before qualitative and quantitative analysis of a sample, and provides reliable and effective data for Raman spectrum substance identification so as to obtain a stable and reliable analysis result. In practical measurements, it is necessary to take measures to suppress fluorescence. In addition, in actual measurement, there are various interference sources in the raman spectrum, mainly including emission noise of laser light and raman scattering light, shot noise, dark current noise and readout noise of a CCD detector, fluorescent and phosphorescent backgrounds of a sample, a sample container, and the like, black body radiation of the sample and its surroundings, and spikes caused by rays in the surroundings, and the like. These sources of interference can cause inaccuracies and instability in the results of subsequent analyses. Therefore, in order to obtain an effective spectrum of an object to be detected, the original spectrum to be detected is subjected to effective fluorescence removal and noise removal treatment, and meanwhile, the spectrum of the object to be detected is subjected to fine characteristic spectrum analysis and practical rapid detection analysis to realize material identification.
With the continuous development of machine learning and artificial intelligence technologies, the automatic and intelligent Raman spectrum data processing and identifying method based on machine learning and the accurate and practical detection tend to be the trend of data processing and identifying development of Raman optical spectrometers.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides the ultraviolet Raman spectrum detection device based on double-probe sampling and adaptive machine learning, which has the advantages of good practicability, quick denoising and wider application range.
In order to achieve the purpose, the ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning comprises the following components:
this ultraviolet raman spectroscopy detection device based on two probe sampling and self-adaptation machine learning, its key feature is, the device include:
the first optical probe is in contact with the surface of the sample through a laser beam and collects Raman signals of the sample, so that fine detection and analysis are realized, and sample information is put into a warehouse;
the second optical probe is a large-aperture remote probe, is in contact with the surface of the sample through a laser beam, collects a Raman signal of the sample, and is complementary with the first optical probe, so that practical rapid detection and analysis are realized;
and the spectrometer is connected with the first optical probe and the second optical probe through optical fibers and is used for receiving the Raman signals of the sample, processing the signals and displaying the signals.
Preferably, the spectrometer comprises:
a processor for controlling and processing the various modules of the spectrometer;
the ultraviolet laser is connected with the processor and used for emitting ultraviolet laser;
a refrigeration detector coupled to the processor for: a semiconductor second-order refrigeration detector is carried, so that the spectrometer has higher wavelength and intensity stability;
the signal preprocessing module is connected with the processor and is used for preprocessing Raman signal data of the sample;
the deep machine learning module is connected with the processor and is used for automatically analyzing and feeding back the Raman signal data of the sample according to a data analysis tool and training a classifier;
and the sample classification and marking module is connected with the processor and used for comparing and identifying the sample types with the database through the deep machine learning module, performing sample type attribution and marking on sample data through the classifier and storing the sample data into the database.
Preferably, the preprocessing operation of the signal preprocessing module includes processing procedures of format specification, data verification, bad data elimination, data smoothing filtering and denoising of the raman signal data.
Preferably, the data analysis tool of the deep machine learning model data processing module includes a principal component analysis unit, a neural network unit, a discriminant function analysis unit and a support vector machine unit.
Preferably, the sample classification and labeling module comprises a classifier, and the classifier is connected with the processor, the signal preprocessing module and the deep machine learning module and is used for data calling.
Preferably, the spectrometer further comprises a display screen connected to the processor for displaying the detection data.
Preferably, the first optical probe has a standard short distance of 7.5mm focal length.
Preferably, the laser focal length of the second optical probe is 1000 mm.
By adopting the ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning, disclosed by the invention, the original Raman data acquired by the Raman spectrometer can be respectively subjected to fine analysis and practical analysis, and automatic and quick de-noising and effective signal identification and classification are realized by combining a self-adaptive machine learning algorithm, so that a quick and effective analysis way is provided for the application of the Raman spectrometer, the application advantages of the Raman spectrometer are reflected, and a quick and effective method is provided for the application of the Raman spectrometer.
Drawings
Fig. 1 is a structural diagram of an ultraviolet raman spectroscopy detection apparatus based on dual-probe sampling and adaptive machine learning according to the present invention.
Detailed Description
In order to more clearly describe the technical contents of the present invention, the following further description is given in conjunction with specific embodiments.
The invention discloses an ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning, which comprises:
the first optical probe is in contact with the surface of the sample through a laser beam and collects Raman signals of the sample, so that fine detection and analysis are realized, and sample information is put into a warehouse;
the second optical probe is a large-aperture remote probe, is in contact with the surface of the sample through a laser beam, collects a Raman signal of the sample, and is complementary with the first optical probe, so that practical rapid detection and analysis are realized;
and the spectrometer is connected with the first optical probe and the second optical probe through optical fibers and is used for receiving the Raman signals of the sample, processing the signals and displaying the signals.
As a preferred embodiment of the present invention, the spectrometer comprises:
a processor for controlling and processing the various modules of the spectrometer;
the ultraviolet laser is connected with the processor and used for emitting ultraviolet laser;
a refrigeration detector coupled to the processor for: a semiconductor second-order refrigeration detector is carried, so that the spectrometer has higher wavelength and intensity stability;
the signal preprocessing module is connected with the processor and is used for preprocessing Raman signal data of the sample;
the deep machine learning module is connected with the processor and is used for automatically analyzing and feeding back the Raman signal data of the sample according to a data analysis tool and training a classifier;
and the sample classification and marking module is connected with the processor and used for comparing and identifying the sample types with the database through the deep machine learning module, performing sample type attribution and marking on sample data through the classifier and storing the sample data into the database.
As a preferred embodiment of the present invention, the preprocessing operation of the signal preprocessing module includes processing procedures of performing format specification, data verification, bad data elimination, data smoothing filtering and denoising on raman signal data.
As a preferred embodiment of the present invention, the data analysis tool of the deep machine learning model data processing module includes a principal component analysis unit, a neural network unit, a discriminant function analysis unit and a support vector machine unit.
As a preferred embodiment of the present invention, the sample classification and labeling module includes a classifier, and is connected to the processor, the signal preprocessing module and the deep machine learning module, and configured to perform data retrieval.
As a preferred embodiment of the present invention, the spectrometer further comprises a display screen connected to the processor for displaying the detection data.
As a preferred embodiment of the invention, the first optical probe has a standard short distance focal length of 7.5 mm.
In a preferred embodiment of the present invention, the focal length of the laser of the second optical probe is 1000 mm.
In a specific embodiment of the invention, the device and the method can realize the original Raman data acquired by the Raman spectrometer by the double optical probes to be refined, have practicability, automatically and quickly remove noise, intelligently identify and classify effective signals, and realize the Raman signal precision data processing and identifying function based on a deep machine learning model.
In order to achieve the purpose, the invention discloses an ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning, which comprises the following steps:
this ultraviolet raman spectroscopy detection device based on two probe sampling and self-adaptation machine learning, its key feature is that the device includes:
the first optical probe is in contact with the surface of the sample through a laser beam, collects Raman signals of the sample, has a standard short-distance focal length of 7.5mm, can realize fine detection and analysis, and stores sample information in a warehouse;
the second optical probe is a large-aperture remote probe, contacts with the surface of a sample through a laser beam, collects a Raman signal of the sample, has a laser focal length of 1000mm, is complementary with the first optical probe, and can realize practical rapid detection and analysis;
the spectrometer is connected with the optical probe and used for receiving the Raman signal of the sample, processing the signal and displaying the signal;
a refrigeration detector: a TEC (semiconductor second-order) refrigeration detector is carried, and the lowest temperature reaches-15 ℃, so that the spectrometer has higher wavelength and intensity stability;
the signal preprocessing module is a component of the spectrometer and is used for preprocessing Raman signal data of a sample;
the self-adaptive machine learning model is a component of the spectrometer and is used for automatically analyzing and feeding back Raman signal data of a sample according to a data analysis tool and training a classifier;
and the sample classification and marking module is a component of the spectrometer phase and is used for performing sample class attribution and marking on sample data through the classifier and storing the sample class attribution and marking in the database.
The laser beam of the optical probe is used for the ultraviolet light.
The spectrometer comprises an ultraviolet laser, a processor, a refrigeration detector, a signal preprocessing module, a self-adaptive deep machine learning model data processing module, a sample classification and labeling module and a display screen. The processor is the brain of the spectrometer and is used for controlling and processing all components of the spectrometer and effectively processing and controlling the signal preprocessing module, the self-adaptive deep machine learning model data processing module, the sample classification and labeling module, the laser and the display screen.
The spectrometer is connected with the optical probe through an optical fiber.
The signal preprocessing module is used for preprocessing the Raman signal data and comprises the steps of carrying out format specification, data verification, bad data elimination, data smooth filtering and denoising on the Raman signal data.
The data analysis tool in the deep machine learning model data processing module comprises a principal component analysis submodule, a neural network submodule, a discriminant function analysis submodule and a support vector machine submodule. The principal component analysis submodule, the neural network submodule, the discriminant function analysis submodule and the support vector machine submodule are all components of the spectrometer.
The sample classification labeling module further comprises a classifier, and the classifier respectively calls data through the processor, the signal preprocessing module and the deep machine learning model data processing module.
The ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning is mainly characterized by comprising the following steps:
(1) the first optical probe and the first optical probe are respectively used for collecting Raman signals for sample refinement and practicability and transmitting the Raman signals to the spectrometer;
(2) a refrigeration detector: a TEC (semiconductor second-order) refrigeration detector is carried, and the lowest temperature reaches-15 ℃, so that the spectrometer has higher wavelength and intensity stability;
(3) the signal preprocessing module is used for preprocessing the Raman signal data of the sample;
(4) the self-adaptive deep machine learning model data processing module analyzes the Raman signal data of the sample according to a data analysis tool;
(5) and the sample classification and labeling module classifies and labels the new sample data through the classifier and stores the new sample data into the database.
The preprocessing of the raman signal data in the step (3) specifically comprises format specification, data verification, bad data elimination, data smooth filtering and denoising of the raman signal data.
The step (5) specifically comprises the following steps:
(5.1) identifying the category of the sample by comparing the sample classification and labeling module with a database through automatic deep learning;
and (5.2) the sample classification and labeling module labels the class of the sample and stores the class in a database.
The ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning comprises a first optical probe, a second optical probe and a spectrometer, wherein the spectrometer further comprises a processor, a refrigeration detector, a laser, a display screen, a signal preprocessing method, a deep machine learning model data processing method and a sample classification and labeling method.
The first optical probe is in contact with the surface of the sample through a laser beam, collects Raman signals of the sample, and can realize fine detection and analysis by a standard short-distance focal length of 7.5 mm;
the second optical probe is a large-aperture remote probe, contacts with the surface of a sample through a laser beam, collects a Raman signal of the sample, has a laser focal length of 1000mm, is complementary with the first optical probe, and can realize practical rapid detection and analysis;
a refrigeration detector: a TEC (semiconductor second-order) refrigeration detector is carried, and the lowest temperature reaches-15 ℃, so that the spectrometer has higher wavelength and intensity stability;
the spectrometer mainly comprises a laser, an optical fiber, a processor, a refrigeration detector, a signal preprocessing module, a deep machine learning model data processing module, a sample classification and labeling module and a display screen. The light source is excited to the [ sample ] to receive Raman signal of the sample, process the signal and display the signal.
A signal preprocessing method is mainly used for carrying out format specification, data verification, bad data elimination, data smooth filtering and denoising on original data, preparing for next machine learning and realizing the preparation before machine learning of the data.
A deep machine learning model data processing method mainly comprises a plurality of data analysis tools such as Principal Component Analysis (PCA), a Neural Network (NN), Discriminant Function Analysis (DFA), a Support Vector Machine (SVM) and the like, is not limited to the enumerated methods, has the characteristics of artificial intelligence in all processes, is fully-automatic and self-adaptive to discriminate, is fully-automatic to process, can display the characteristic spectrum of a substance, can not display the characteristic spectrum, and can identify the substance directly by comparing the deep automatic learning with a database.
A classifier trained by machine learning and data processing is utilized to perform sample class attribution and labeling on input new sample data, and the input new sample data is stored in a database.
The ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning designed and realized according to the invention provides a function of quickly and intelligently analyzing material characteristics for application of a Raman optical spectrometer. The application advantage of the Raman spectrometer is reflected.
As shown in fig. 1, the present embodiment includes a first optical probe, a second optical probe, an optical fiber, and a spectrometer, wherein the spectrometer further includes a laser, an optical fiber, a processor, a refrigeration detector, a signal preprocessing module, a deep machine learning model data processing module, a sample classification labeling module, and a display screen.
The first optical probe and the first optical probe are respectively connected with the spectrometer through optical fibers, on one hand, the optical probe obtains a light source from a laser of the spectrometer and collects Raman signals of a sample, and then the optical probe transmits the collected Raman original signals of the sample to the spectrometer through the optical fibers; the spectrometer mainly comprises a laser, a processor, a refrigeration detector, a signal preprocessing module, a deep machine learning model data processing module, a sample classification marking module and a display screen, and is used for exciting a light source for a sample, receiving Raman signals of an optical probe, processing the signals and displaying the signals; the signal preprocessing mainly comprises data format specification, data verification and bad data elimination, data smooth filtering and denoising, and the preparation work before the machine learning of the data is realized;
the deep machine learning model data processing mainly adopts a plurality of data analysis tools to automatically analyze data, the tools comprise Principal Component Analysis (PCA), a Neural Network (NN), Discriminant Function Analysis (DFA), a Support Vector Machine (SVM) and the like, and all processes have the characteristic of artificial intelligence and are processed fully automatically. The sample classification and labeling utilizes a classifier trained by machine learning and data processing to perform new sample class attribution and labeling on input new sample data, and the new sample class attribution and labeling are stored in a database, so that the characteristic spectrum of a substance can be displayed, or the characteristic spectrum can not be displayed, and the substance can be identified by directly comparing the deep automatic learning with the database.
By adopting the ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning, disclosed by the invention, the original Raman data acquired by the Raman spectrometer can be respectively subjected to fine analysis and practical analysis, and automatic and quick de-noising and effective signal identification and classification are realized by combining a self-adaptive machine learning algorithm, so that a quick and effective analysis way is provided for the application of the Raman spectrometer, the application advantages of the Raman spectrometer are reflected, and a quick and effective method is provided for the application of the Raman spectrometer.
In this specification, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Claims (8)
1. An ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning is characterized by comprising:
the first optical probe is in contact with the surface of the sample through a laser beam and collects Raman signals of the sample, so that fine detection and analysis are realized, and sample information is put into a warehouse;
the second optical probe is a large-aperture remote probe, is in contact with the surface of the sample through a laser beam, collects a Raman signal of the sample, and is complementary with the first optical probe, so that practical rapid detection and analysis are realized;
and the spectrometer is connected with the first optical probe and the second optical probe through optical fibers and is used for receiving the Raman signals of the sample, processing the signals and displaying the signals.
2. The apparatus for ultraviolet-raman spectroscopy detection based on dual-probe sampling and adaptive machine learning according to claim 1, wherein the spectrometer comprises:
a processor for controlling and processing the various modules of the spectrometer;
the ultraviolet laser is connected with the processor and used for emitting ultraviolet laser;
a refrigeration detector coupled to the processor for: a semiconductor second-order refrigeration detector is carried, so that the spectrometer has higher wavelength and intensity stability;
the signal preprocessing module is connected with the processor and is used for preprocessing Raman signal data of the sample;
the deep machine learning module is connected with the processor and is used for automatically analyzing and feeding back the Raman signal data of the sample according to a data analysis tool and training a classifier;
and the sample classification and marking module is connected with the processor and used for comparing and identifying the sample types with the database through the deep machine learning module, performing sample type attribution and marking on sample data through the classifier and storing the sample data into the database.
3. The UV-Raman spectrum detection apparatus based on dual-probe sampling and adaptive machine learning of claim 2, wherein the preprocessing operation of the signal preprocessing module comprises processing procedures of format specification, data verification, bad data rejection, data smoothing filtering and denoising for Raman signal data.
4. The UV-Raman spectrum detection apparatus based on dual-probe sampling and adaptive machine learning according to claim 2, wherein the data analysis tool of the deep machine learning model data processing module comprises a principal component analysis unit, a neural network unit, a discriminant function analysis unit and a support vector machine unit.
5. The ultraviolet-raman spectroscopy detection device based on dual-probe sampling and adaptive machine learning of claim 2, wherein the sample classification and labeling module comprises a classifier connected with the processor, the signal preprocessing module and the deep machine learning module for data calling.
6. The UV-Raman spectrum detection apparatus based on dual-probe sampling and adaptive machine learning of claim 2, wherein the spectrometer further comprises a display screen connected to the processor for displaying detection data.
7. The UV-Raman spectrum detection apparatus based on dual-probe sampling and adaptive machine learning of claim 1, wherein the first optical probe has a standard short-range 7.5mm focal length.
8. The UV-Raman spectrum detection apparatus based on dual-probe sampling and adaptive machine learning of claim 1, wherein a laser focal length of the second optical probe is 1000 mm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010777408.5A CN111879752A (en) | 2020-08-05 | 2020-08-05 | Ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010777408.5A CN111879752A (en) | 2020-08-05 | 2020-08-05 | Ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111879752A true CN111879752A (en) | 2020-11-03 |
Family
ID=73210551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010777408.5A Pending CN111879752A (en) | 2020-08-05 | 2020-08-05 | Ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111879752A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1149334A (en) * | 1994-05-27 | 1997-05-07 | 伊斯特曼化学公司 | Raman spectrometry apparatus and method |
| CN1981189A (en) * | 2004-07-02 | 2007-06-13 | 皇家飞利浦电子股份有限公司 | Spectroscopic system with multiple probes |
| CN101614667A (en) * | 2008-06-27 | 2009-12-30 | 同方威视技术股份有限公司 | Raman spectrum system and method for measuring Raman spectrum |
| CN103645170A (en) * | 2013-12-03 | 2014-03-19 | 北京航空航天大学 | Device for detecting tumor characters by Raman spectrum |
| CN104713865A (en) * | 2013-12-13 | 2015-06-17 | 中国科学院大连化学物理研究所 | Deep ultraviolet laser Raman spectrometer |
| CN107192703A (en) * | 2017-07-06 | 2017-09-22 | 安徽科创中光科技有限公司 | A kind of Pollution From Ships discharge remote sensing detection system |
| CN109190714A (en) * | 2018-10-11 | 2019-01-11 | 公安部第三研究所 | The system and method that Raman signal identifies is realized based on depth machine learning model |
| CN209117581U (en) * | 2018-10-23 | 2019-07-16 | 高利通科技（深圳）有限公司 | A kind of combination Raman spectrum analysis system |
-
2020
- 2020-08-05 CN CN202010777408.5A patent/CN111879752A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1149334A (en) * | 1994-05-27 | 1997-05-07 | 伊斯特曼化学公司 | Raman spectrometry apparatus and method |
| CN1981189A (en) * | 2004-07-02 | 2007-06-13 | 皇家飞利浦电子股份有限公司 | Spectroscopic system with multiple probes |
| CN101614667A (en) * | 2008-06-27 | 2009-12-30 | 同方威视技术股份有限公司 | Raman spectrum system and method for measuring Raman spectrum |
| CN103645170A (en) * | 2013-12-03 | 2014-03-19 | 北京航空航天大学 | Device for detecting tumor characters by Raman spectrum |
| CN104713865A (en) * | 2013-12-13 | 2015-06-17 | 中国科学院大连化学物理研究所 | Deep ultraviolet laser Raman spectrometer |
| CN107192703A (en) * | 2017-07-06 | 2017-09-22 | 安徽科创中光科技有限公司 | A kind of Pollution From Ships discharge remote sensing detection system |
| CN109190714A (en) * | 2018-10-11 | 2019-01-11 | 公安部第三研究所 | The system and method that Raman signal identifies is realized based on depth machine learning model |
| CN209117581U (en) * | 2018-10-23 | 2019-07-16 | 高利通科技（深圳）有限公司 | A kind of combination Raman spectrum analysis system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7701573B2 (en) | Method for correlating spectroscopic measurements with digital images of contrast enhanced tissue | |
| Dong et al. | Automated analysis of microplastics based on vibrational spectroscopy: are we measuring the same metrics? | |
| CN109781622B (en) | Portable intelligent metal ore type quick distinguishing instrument | |
| CN107860473A (en) | A kind of intelligent handhold spectrometer based on cloud data | |
| US11879829B2 (en) | Methods and systems for classifying fluorescent flow cytometer data | |
| CN113155809B (en) | Novel spectral detection method for ore classification and real-time quantitative analysis | |
| US9002113B2 (en) | Processing and analyzing hyper-spectral image data and information via dynamic database updating | |
| CN111707656B (en) | Cerebrospinal fluid cell detection method and system based on Raman scattering spectrum | |
| CN109190714A (en) | The system and method that Raman signal identifies is realized based on depth machine learning model | |
| CN111033230A (en) | Real-time spectral analysis using high-speed classification of spectra | |
| CN116030310A (en) | Sample classification method and system based on laser-induced breakdown spectroscopy technology | |
| CN113514446B (en) | Method for rapidly matching and identifying SERS spectrogram | |
| CN106198481B (en) | Fake wine identification device and method based on LIF technology and naive Bayes classification | |
| US20210278333A1 (en) | Methods and systems for adjusting a training gate to accommodate flow cytometer data | |
| CN111879752A (en) | Ultraviolet Raman spectrum detection device based on double-probe sampling and self-adaptive machine learning | |
| US8289503B2 (en) | System and method for classifying a disease state using representative data sets | |
| US11662295B2 (en) | Deep learning method in aiding patient diagnosis and aberrant cell population identification in flow cytometry | |
| CN113466206A (en) | Raman spectrum analysis system based on big data | |
| CN209070559U (en) | Support the system structure of realization Raman signal data processing and identification function | |
| US20210270718A1 (en) | Methods for identifying saturated data signals in cell sorting and systems for same | |
| Oßmann | Microplastics characterization by Raman microscopy | |
| CN112683859B (en) | Component quantitative analysis method, test system and storage medium | |
| US11231369B2 (en) | Portable single unit device for Ochratoxin A (OTA) toxicity analysis for rice quality monitoring | |
| CN115171805A (en) | System for resolved biomolecular mass spectrometry imaging and methods of use thereof | |
| US20220155209A1 (en) | Method for Optimal Scaling of Cytometry Data for Machine Learning Analysis and Systems for Same |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201103 |
|
| RJ01 | Rejection of invention patent application after publication |