CN107991283A - Raman spectrum detection device and Raman spectra detection process - Google Patents
Raman spectrum detection device and Raman spectra detection process Download PDFInfo
- Publication number
- CN107991283A CN107991283A CN201711437491.6A CN201711437491A CN107991283A CN 107991283 A CN107991283 A CN 107991283A CN 201711437491 A CN201711437491 A CN 201711437491A CN 107991283 A CN107991283 A CN 107991283A
- Authority
- CN
- China
- Prior art keywords
- image
- danger zone
- predetermined danger
- face
- raman spectrum
- 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.)
- Granted
Links
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
Landscapes
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (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)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The embodiment provides a kind of Raman spectrum detection device and a kind of Raman spectra detection process.Raman spectrum detection device includes：Exciting light sources, are configured to launch exciting light to sample to be tested；Optical devices, have spectral detection light path and a predetermined danger zone imaging optical path, spectral detection light path be configured to collect from sample to be tested be excited light irradiation position optical signal, predetermined danger zone imaging optical path is configured to shoot the image of predetermined danger zone；Spectrometer, is configured to receive the optical signal from spectral detection light path and by the Raman spectrum of optical signal generation sample to be tested；Face identification device, is configured to receive the image of predetermined danger zone and identifies in the image of predetermined danger zone whether include face；And safety governor, it is configured to receive the recognition result of face identification device and exciting light sources is off state in the case where face identification device identifies that the image of predetermined danger zone includes face.
Description
Technical field
The embodiment of the present invention is related to Raman spectrum detection field, more particularly to a kind of Raman spectrum detection device and one kind
Raman spectra detection process.
Background technology
Raman spectrum analysis technology is a kind of contactless spectral analysis technique based on Raman scattering effect, its energy
Qualitative and quantitative analysis is carried out to the component of material.Raman spectrum is a kind of molecular vibration spectrum, it can reflect the fingerprint of molecule
Feature, available for the detection to material.Raman spectrum is detected by detecting Raman scattering effect institute of the determinand for exciting light
The Raman spectrum of generation detects and identifies material.Raman spectra detection process has been widely used for liquid safety check, jewelry inspection
The fields such as survey, explosive detection, illicit drugs inspection, medicine detection, Detecting Pesticide.
In recent years, Raman spectrum analysis technology is widely used in the field such as dangerous material inspection and Object Classification.
In Object Classification field, due to the color of various materials, different, people usually can not accurate judgement thing qualitative attribution, and
Raman spectrum is by the molecular entergy level structures shape of checking matter, thus Raman spectrum can be as " fingerprint " information of material, for thing
Matter identifies.Therefore Raman spectrum analysis technology is widely used in fields such as customs, public safety, food and medicine, environment.
Since Raman spectrum generally requires, by the use of the laser of high power density as excitation source, may have stronger thermal effect
Should, in the case where sample is unknown, detection rashly is possible to sample can be caused to be damaged by laser ablation, it could even be possible to causing to swash
Light ignites or ignites some inflammable and explosive chemicals, causes the loss of personal property.Moreover, in some cases, such as Raman
The exciting light that spectrum detection instrument is sent may also hurt human eye.
The content of the invention
The present invention is directed to propose a kind of Raman spectrum detection device and a kind of Raman spectra detection process, it can be effective
Ground reduces or avoids the risk that personnel come to harm in Raman spectrum detects.
The embodiment provides a kind of Raman spectrum detection device, including：Exciting light sources, are configured to to be measured
Electromagnetic radiation exciting light；Optical devices, the Optical devices have spectral detection light path and predetermined danger zone imaging optical path, institute
State spectral detection light path be configured to collect from the sample to be tested be excited light irradiation position optical signal, the predetermined danger
Danger zone domain imaging light path is configured to shoot the image of predetermined danger zone；Spectrometer, is configured to reception and comes from the spectral detection
The optical signal of light path and the Raman spectrum that sample to be tested is generated by the optical signal；Face identification device, is configured to described in reception
The image of predetermined danger zone simultaneously identifies in the image of the predetermined danger zone whether include face；And safety governor,
It is configured to receive the recognition result of face identification device and identifies the image of the predetermined danger zone in face identification device
Include making exciting light sources be off state in the case of face.
In one embodiment, the predetermined danger zone includes be excited light irradiation or the region by the light irradiation that is excited.
In one embodiment, the predetermined danger zone includes the surrounding environment area of the Optical devices or sample to be tested
Domain.
In one embodiment, the spectral detection light path includes collecting lens, spectroscope, Raman filter set and coupling successively
Lens are closed, and predetermined danger zone imaging optical path includes collecting lens, spectroscope, lens group and image taking instrument, wherein, institute
State collecting lens and spectroscope be the spectral detection light path and predetermined danger zone imaging optical path common unit.
In one embodiment, spectral detection light path and predetermined danger zone imaging optical path are separated from each other.
In one embodiment, the face identification device includes：Image characteristics extraction module, described image feature extraction mould
Block is configured to the feature of extraction described image and generates feature vector；Sorter model generation module, the sorter model life
It is configured into module according to reference to Face image synthesis sorter model；And identification module, the identification module are configured to institute
Feature vector is stated as the input vector of the sorter model to judge what is collected by predetermined danger zone imaging optical path
Whether face characteristic is included in image.
The embodiment of the present invention additionally provides a kind of Raman spectra detection process, including：Shoot the figure of predetermined danger zone
Picture；Whether identify in the image of the predetermined danger zone includes face；And in the figure for identifying the predetermined danger zone
As including making exciting light sources be off state in the case of face, and in the image for identifying the predetermined danger zone
In do not include face in the case of, start exciting light sources, gather the Raman spectrum of sample to be detected to sample.
In one embodiment, it is described before whether including face in the image of the identification predetermined danger zone
Method further includes：According to reference to Face image synthesis sorter model.
In one embodiment, whether include in the image of the identification predetermined danger zone including face：Extraction institute
State the feature of image and generate feature vector；With using described eigenvector as the input vector of the sorter model to judge
Whether face characteristic is included in the image collected by predetermined danger zone imaging optical path.
Examined by means of the secure sample detected according to the Raman spectrum detection device and Raman spectrum of above-described embodiment
Survey method, can cut down or prevent during spectral detection because exciting light is ignited, examined caused by ablation or ignition sample
Survey risk.
Brief description of the drawings
Fig. 1 shows the schematic diagram of the Raman spectrum detection device of an embodiment according to the present invention；
Fig. 2 shows the schematic diagram of the Raman spectrum detection device of an embodiment according to the present invention；
Fig. 2 a show the predetermined danger zone imaging in the Raman spectrum detection device of an embodiment according to the present invention
The schematic diagram of the equivalent light path of light path；
Fig. 3 shows the flow chart of the Raman spectra detection process of an embodiment according to the present invention；
Fig. 4 shows that what image feature vector in the Raman spectra detection process of an embodiment according to the present invention extracted shows
The flow chart of example property；
Fig. 5 shows the example that sorter model is established in the Raman spectra detection process of an embodiment according to the present invention
The flow chart of property；And
Fig. 6 shows the exemplary stream of feature extraction in the Raman spectra detection process of an embodiment according to the present invention
Cheng Tu.
Attached drawing not to all circuits in Raman spectrum detection device according to an embodiment of the invention or structure into
Row display.The reference numeral identical through all attached drawings represents the same or similar component or feature.
Embodiment
Below with reference to the embodiments and with reference to the accompanying drawing the technical solutions of the present invention will be further described.Illustrating
In book, the same or similar drawing reference numeral represents the same or similar component.It is following referring to the drawings to embodiment of the present invention
Illustrate to be intended to explain the present general inventive concept of the present invention, and be not construed as limiting a kind of of the present invention.
In addition, in the following detailed description, for ease of explaining, many concrete details are elaborated to provide to present disclosure
The comprehensive understanding of embodiment.It should be apparent, however, that one or more embodiments also may be used in the case of these no details
To be carried out.In other cases, known construction and device is diagrammatically embodied to simplify attached drawing.
The embodiment provides a kind of Raman spectrum detection device 100.As shown in Figure 1, the Raman spectrum detects
Equipment 100 includes：Exciting light sources 10, Optical devices 20, spectrometer 30, face identification device 40 and safety governor 60.Should
Exciting light sources 10, such as laser can be included, it is configured to launch exciting light to sample to be tested 50.The Optical devices 20 can
With including spectral detection light path 21 and predetermined danger zone imaging optical path 22.The spectral detection light path 21 is configured to collection and comes from
The sample to be tested 50 be excited light irradiation position optical signal.Collected optical signal can be transferred into spectrometer
30.Spectrometer 30, which then can be configured to, to be received the optical signal from the spectral detection light path 21 and is generated by the optical signal to be measured
The Raman spectrum of sample 50, so as to fulfill the normal flow of Raman spectrum detection.And the predetermined danger zone imaging optical path 22
It is configured to shoot be excited light irradiation or the image by the predetermined danger zone for the light irradiation that is excited, which is transmitted to face
Identification device 40.The face identification device 40 then can be configured to based on the institute shot by the predetermined danger zone imaging optical path 22
The image of predetermined danger zone is stated to determine whether there is face to exist in predetermined danger zone.
As it was previously stated, using exciting light come to gather optical signalling be basic step that Raman spectrum detection is carried out to sample to be tested
Suddenly, and exciting light has certain energy in itself, for the sample to be tested of some materials, it may occur anti-with exciting light
Answer and cause the change of sample composition.For example, some inflammable, explosive materials may be ignited under the action of exciting light, burn
Erosion, explosion etc..In practice, the component of sample to be tested is often unknown, so if the face of people, which is in distance, treats test sample
The nearer position of product (such as surrounding area of Optical devices or sample to be tested), it is likely that come to harm.For another example swash
Shine itself may also produce injury to human eye, therefore, can also be by detecting in quilt before excitation light emission or in transmitting
Exciting light irradiates or may be excited on the region of light irradiation the face that whether there is people, to avoid risk.This is for hand-held
Raman spectrum detection device is even more important.
In Raman spectrum detection device according to an embodiment of the invention, safety governor 60 can be configured to recipient
The recognition result of face identification device 40 simultaneously identifies that the image of the predetermined danger zone includes people in face identification device 40
Exciting light sources are made to be off state in the case of face.In this manner it is possible to prevent from that personnel's injury occurs in detection process.
In one example, as shown in Figure 1, spectral detection light path 21 can include collecting lens 31, spectroscope 32 successively, draw
Graceful filter set 33 and coupled lens 34.As an example, Raman filter set 33 can include one or more optical filters, use
Retain Raman scattering optical signal in the filtering undesirable light such as Rayleigh scattering light and exciting light.Coupled lens 34 can be used for by
The optical signal filtered by Raman filter set 33 is coupled to spectrometer 30.Predetermined danger zone imaging optical path 22 can include saturating
Microscope group 23 and image taking instrument 24.The image taking instrument 24 such as can be camera (as CCD camera), shooting it is first-class
Device for shooting image.Lens group 23 can include one or more lens, it can be and image taking instrument 24
Discrete, it can also be integrated with image taking instrument 24.The lens group 23 can be made of any number of lens,
It can be made of any lens group that can realize blur-free imaging known in the art.The predetermined danger zone shown in Fig. 1 into
As light path 22 is schematical, can select to wish the predetermined danger zone being imaged as needed." predetermined danger mentioned here
Danger zone domain " refers to that the face of the people in optical excitation signal gatherer process is in the region area that potential danger may occur
Domain.For example, the predetermined danger zone can include be excited light irradiation or the region by the light irradiation that is excited, or including described
The surrounding area of Optical devices or sample to be tested.The surrounding area of Optical devices or sample to be tested refers in the region
In position and Optical devices or the distance of sample to be tested be less than safe distance (safe distance can be according to the point that may occur
Combustion, explosion, the energy of laser emission determine).Recognition of face is carried out by the image for these predetermined danger zones, can be with
Determine whether there is the risk for causing personnel to injure.
As an example, when being imaged in the environment of dark, the devices such as light compensating lamp may be it is also required to provide and come for shooting
The image of sample to be tested provides sufficient intensity of illumination to improve the clarity of image.
In the example shown in above-mentioned Fig. 1, spectral detection light path 21 and predetermined danger zone imaging optical path 22 are to divide each other
From.And in another example, there may also be certain friendship for spectral detection light path 21 and predetermined danger zone imaging optical path 22
It is folded.For example, as shown in Fig. 2, collecting lens 31 and spectroscope 32 are spectral detection light path 21 and predetermined danger zone imaging optical path
22 common unit.This is conducive to the compactness for improving Optical devices, to save space.
In the Raman spectrum detection device 100 ' according to another embodiment of the present invention shown in Fig. 2, spectral detection light
Road 21 and predetermined danger zone imaging optical path 22 have common portion at sample to be tested 50, and spectroscope 32, which allows to come from, to be treated
The optical signal of sample 50 is transmitted through with directive spectrometer, so as to form spectral detection light path 21, spectroscope 32 will carry figure
The light beam of picture is reflected to form predetermined danger zone imaging optical path 22.It should be noted that although spectral detection light path 21 and pre-
Determining danger zone imaging optical path 22 has common portion, still, it is not intended that spectral detection and to predetermined danger zone be imaged
Operation to be carried out at the same time, for example, to predetermined danger zone imaging operation carried out usually before spectral detection, with spectrum
Security risk is excluded before detection, it is dangerous to avoid occurring during spectral detection.In addition, spectral detection light path 21 and predetermined danger
Danger zone domain imaging light path 22 shares collecting lens 31, and the region that may preferably can be irradiated for exciting light is imaged.
It should be noted that Fig. 2 is illustrated that in Optical devices 20 in the case that exciting light sources 10 launch exciting light
Light beam traveling situation, however, in practice, when being imaged for predetermined danger zone, often exciting light sources are simultaneously
Do not launch exciting light, even there are sample to be tested 50.At this time, the equivalent light path of predetermined danger zone imaging optical path 22 such as Fig. 2 a
Shown, collecting lens 31 and lens group 23 can project the image of its towards the side of sample to be tested 50 to image taking instrument
24, and image is imaged in image plane (such as plane where egative film, CCD etc.) by the camera lens 26 of image taking instrument 24.This
The image in the region in the front of collecting lens 31 can be easily obtained, this can be located at the check bit of sample to be tested 50 to avoid human eye
Risk caused by putting nearby.
As an example, the light path that exciting light sources 10 launch exciting light can also partially overlap with spectral detection light path 21,
For example, in the example shown in Fig. 1 and Fig. 2, another spectroscope 35 can be set, which can be by exciting light sources 10
The excitation light reflection sent, the exciting light reflected are focused on sample to be tested 50 by collecting lens 31, this helps to simplify
The adjustment of light path, however, the embodiment of the present invention not limited to this, such as the light that sends of exciting light sources 10 can pass through and spectrum
Light path completely self-contained light path irradiates sample to be tested.As an example, exciting light may be used also after being sent from exciting light sources 10
By removing veiling glare optical filter 36, should can be used for removing veiling glare except veiling glare optical filter 36 to improve the signal-to-noise ratio of exciting light.
Above-mentioned example is only to give a kind of exemplary implementation of Raman spectrum detection device, but the present invention
Embodiment not limited to this, the Raman spectrum detection that other those skilled in the art anticipated that after present disclosure is read
The alternative of equipment is also feasible.
As an example, face identification device 40 can include：Image characteristics extraction module 41, described image feature extraction mould
Block is configured to the feature of extraction described image and generates feature vector；Sorter model generation module 42, the sorter model
Generation module is configured to according to reference to Face image synthesis sorter model；With identification module 43, the identification module is configured to
Using described eigenvector as the input vector of the sorter model to judge to be gathered by predetermined danger zone imaging optical path
To image in whether include face characteristic.
In the above-described embodiments, face identification device 40 and safety governor 60 can be realized by processor, can also
Realized by other soft-hardware configurations.And as an example, safety governor 60 can also be realized by trigger switch.
The embodiment of the present invention also discloses a kind of Raman spectra detection process S100.As shown in figure 3, this method can wrap
Include：
Step S10：Shoot the image of predetermined danger zone；
Step S20：Whether identify in the image of the predetermined danger zone includes face；And
Step S30：Make exciting light sources in the case where the image for identifying the predetermined danger zone includes face
State is off, and in the case of not including face in the image for identifying the predetermined danger zone, start exciting light
Light source, gathers the Raman spectrum of sample to be detected to sample.
Referring to the step of whether including face in images of the Fig. 4 to Fig. 6 for identifying the predetermined danger zone into
The exemplary introduction of row：
Fig. 4 shows the example of image characteristics extraction flow., it is necessary to have training sample first for recognition of face
This, that is to say, that the known image with face, hereon referred to as with reference to facial image.Image is extracted by the training sample
Feature, form sampling feature vectors, can establish sorter model, and is adopted in practice by predetermined danger zone imaging optical path 22
The image collected can also equally be extracted the feature of image, form the feature vector of test image, and be based on the grader mould
Type calculates the matching degree with the model, with determine actual acquisition to test image in whether there is face characteristic.
In flow chart illustrated in fig. 4, including the step of step S41, i.e. collection training sample.Gathering training sample can
To carry out Image Acquisition at random by Raman spectrum detection device according to an embodiment of the invention, but in the case, adopted
Need to include face characteristic in the training sample collected, that is to say, that when carrying out Image Acquisition to predetermined danger zone, make a reservation for
There should be face to be located in it in danger zone.Alternatively, gathering training sample can also be by from known (facial image)
Directly acquired in sample storehouse.
After training sample image is collected, step S42 is performed, that is, carries out the extraction of feature.As an example, it can scheme
As in characteristic extraction procedure, selecting image dense feature.In this example, use field of image recognition have superior function and
The SIFT feature being widely used (SIFT feature is known in the art, and details are not described herein).Characteristic extraction step is as follows, such as
Shown in Fig. 6：
1. the number of the training sample image collected is counted, in this example, which is N1。
2. determining the stepping length lambda in characteristic extraction procedure, sliding forms sizes, (wherein λ and α are nature for α
Number)；
3. training sample image k is inputted, including the length (i.e. the number of lines of pixels (img.cols) of image) of record image, width
(i.e. the pixel columns (img.rows) of image) is respectively m, n, juxtaposition i=0, j=0；
4. extracting the SIFT feature that (i+ α, j+ α) image block (patch) is arrived in image (i, j), record SIFT feature is f；
5. j is judged whether j+ α≤n set up from stepping length lambda is increased.If set up repeat 2. 3. 4. the step of, otherwise jump
Go to step 6.；
6. by j zero setting, i judges whether j+ α≤m set up from stepping length lambda is increased.Step 2. 3. 4. is repeated if setting up
Suddenly, until i>M completes dense feature (SIFT feature) F of training sample image kk=[f1,f2,…,fNumk] extraction, wherein
Numk=((n- α)/λ) * ((m- α)/λ), k=1,2 ..., N1。
After dense feature extraction is completed, then need to perform step S43, i.e. dictionary construction and feature coding.Dictionary structure
The process of making can include：
Randomly select the SIFT feature F=[F of training sample image1,F2,…,FN1], wherein Fk=[f1,f2,…,fNumk],
It is recorded as M=[f1,f2,…,fm], wherein choosing
Clustering processing is carried out using characteristics of image M, it is h to choose cluster centre number, is such as gathered using known K-means
Class algorithm, obtains eigencenter vector, so as to generate dictionary D, D=[d1,d2,...,dh]。
Feature coding flow can include：
Input the intensive SIFT feature F of training sample image kk=[f1,f2,…,fNumk]；
(1) carries out reconstruction from projection to the SIFT feature of training sample image k according to the following equation, records ci, that is, train sample
The feature vector of this image, that is, grader input vector.
Wherein, " s.t. " represents constraints, for N1For a training sample image, N=N1。
In order to establish sorter model, using the feature vector of above-mentioned training sample image as input sample, can select
Linear SVM model known in the art is trained as grader, obtains the sorter model of training sample, such as
Shown in Fig. 5.Training sample label refers to the label corresponding to training sample image in Fig. 5.
As an example, in order to preferably confirm the correctness of sorter model, model verification can be carried out.In order to which model is tested
Card, can utilize the model similar with training sample image to verify sample image, according to the spy of said extracted training sample image
The step of vector is levied, dense feature extraction is carried out to model verification sample image, characteristics dictionary is projected, to form model
Verify the feature vector of sample image.The feature vector of model verification sample image may be used as the defeated of sorter model verification
Incoming vector, whether verification classification output is correct, and the recall rate of the accuracy rate, i.e. model of statistical testing of business cycles output.As an example, instruction
The quantity and the ratio of number of model verification sample image for practicing sample image can be 5：1.It will be appreciated by those skilled in the art that
The illustrative steps that above-mentioned model verification flow provides just for the sake of optimization algorithm, the embodiment of the present invention can not include
The model verifies flow.
As an example, whether include face in the real image for judging to be collected by predetermined danger zone imaging optical path 22
When, the feature vector of the real image first can be extracted (such as according to above-mentioned image feature vector extraction flow as shown in Figure 4
Including extraction dense feature, projection dictionary, feature coding etc.), then using the feature vector of real image as input vector band
Enter in above-mentioned sorter model, to judge its matching degree with sorter model, that is, judge whether the real image includes people
Face.
It is merely exemplary it will be appreciated by those skilled in the art that above-mentioned on the step of recognition of face, it is of the invention
Embodiment not limited to this, known in the art other can realize that the method for recognition of face is used equally for the embodiment of the present invention.
In the Raman detection method of an embodiment according to the present invention, the figure in the identification predetermined danger zone
Before whether including face as in, the method can also include：
Step S00：According to reference to Face image synthesis sorter model.
Specific exemplary flow has been presented above, and details are not described herein.
In the Raman detection method of an embodiment according to the present invention, above-mentioned steps S20 can include：
Step S21：Extract the feature of described image and generate feature vector；With
Step S22：Using described eigenvector as the input vector of the sorter model to judge by predetermined hazardous area
Whether face characteristic is included in the image that domain imaging light path is collected.
Specific exemplary flow has been presented above, and details are not described herein.
In the image of the identification predetermined danger zone in above-mentioned Raman detection method according to an embodiment of the invention
The step of whether including face, can correspondingly be filled by the recognition of face in Raman detection equipment according to an embodiment of the invention
40 modules are put to perform.
Detailed description above has elaborated above-mentioned Raman spectrum inspection by using schematic diagram, flow chart and/or example
Numerous embodiments of measurement equipment and method.This schematic diagram, flow chart and/or example include one or more functions and/or
In the case of operation, it will be understood by those skilled in the art that each function and/or behaviour in this schematic diagram, flow chart or example
Making can be by various structures, hardware, software, firmware or substantially their any combination come independent and/or common realization.
In one embodiment, if the stem portion of theme described in the embodiment of the present invention can pass through application-specific integrated circuit (ASIC), scene
Programmable gate array (FPGA), digital signal processor (DSP) or other integrated forms are realized.However, people in the art
Member is it should be understood that some aspects of embodiments disclosed herein can be realized equally in integrated electricity on the whole or partly
Lu Zhong, the one or more computer programs for being embodied as running on one or more computer (for example, be embodied as at one or
The one or more programs run in multiple stage computers system), be embodied as running on the one or more processors one or
Multiple programs (for example, being embodied as the one or more programs run in one or more microprocessors), are embodied as firmware, or
Person is substantially embodied as any combination of aforesaid way, and those skilled in the art will be provided with design circuit according to the disclosure
And/or the ability of write-in software and/or firmware code.In addition, it would be recognized by those skilled in the art that theme described in the disclosure
Mechanism can be distributed as the program product of diversified forms, and no matter actually be used for performing the signal bearing medium of distribution
Particular type how, the exemplary embodiment of theme described in the disclosure is applicable in.The example of signal bearing medium is included but not
It is limited to：Recordable-type media, such as floppy disk, hard disk drive, CD (CD, DVD), digital magnetic tape, computer storage；And
Transmission type media, such as numeral and/or analogue communication medium are (for example, optical fiber cable, waveguide, wired communications links, wireless communication link
Road etc.).
Unless there are technology barrier or contradiction, above-mentioned various embodiments of the invention can be freely combined to be formed in addition
Embodiment, these further embodiments are in protection scope of the present invention.
Although with reference to attached drawing, the present invention is described, and the embodiment disclosed in attached drawing is intended to preferred to the present invention
Embodiment is illustrative, and it is not intended that limiting a kind of of the present invention.Dimension scale in attached drawing is only
Schematically, limitation of the present invention can not be interpreted as.
Although some embodiments of present general inventive concept are shown and illustrate, those of ordinary skill in the art will manage
Solution, in the case of without departing substantially from the principle of this present general inventive concept and spirit, can make a change these embodiments, of the invention
Scope is limited with claim and their equivalent.
Claims (9)
1. a kind of Raman spectrum detection device, including：
Exciting light sources, are configured to launch exciting light to sample to be tested；
Optical devices, the Optical devices have spectral detection light path and predetermined danger zone imaging optical path, the spectral detection
Light path be configured to collect from the sample to be tested be excited light irradiation position optical signal, the predetermined danger zone imaging
Light path is configured to shoot the image of predetermined danger zone；
Spectrometer, is configured to receive the optical signal from the spectral detection light path and by optical signal generation sample to be tested
Raman spectrum；
Face identification device, is configured to receive the image of the predetermined danger zone and identifies the image of the predetermined danger zone
In whether include face；And
Safety governor, is configured to receive the recognition result of face identification device and is identified in face identification device described predetermined
The image of danger zone includes making exciting light sources be off state in the case of face.
2. Raman spectrum detection device according to claim 1, wherein, the predetermined danger zone includes the illumination that is excited
The region penetrated or irradiate the light that is excited.
3. Raman spectrum detection device according to claim 1, wherein, the predetermined danger zone is filled including the optics
Put or the surrounding area of sample to be tested.
4. Raman spectrum detection device according to claim 1, wherein, the spectral detection light path includes collecting successively saturating
Mirror, spectroscope, Raman filter set and coupled lens, and predetermined danger zone imaging optical path include collecting lens, spectroscope, thoroughly
Microscope group and image taking instrument, wherein, the collecting lens and spectroscope are the spectral detection light path and predetermined danger zone
The common unit of imaging optical path.
5. Raman spectrum detection device according to claim 1, wherein, spectral detection light path and the imaging of predetermined danger zone
Light path is separated from each other.
6. Raman spectrum detection device according to any one of claim 1 to 5, wherein, the face identification device bag
Include：
Image characteristics extraction module, described image characteristic extracting module be configured to extraction described image feature and generate feature to
Amount；
Sorter model generation module, the sorter model generation module are configured to according to reference to Face image synthesis grader
Model；With
Identification module, the identification module are configured to using described eigenvector as the input vector of the sorter model to sentence
Whether face characteristic is included in the disconnected image collected by predetermined danger zone imaging optical path.
7. a kind of Raman spectra detection process, including：
Shoot the image of predetermined danger zone；
Whether identify in the image of the predetermined danger zone includes face；And
Exciting light sources are made to be off state in the case where the image for identifying the predetermined danger zone includes face,
And in the case of not including face in the image for identifying the predetermined danger zone, start exciting light sources, gather sample
Raman spectrum to be detected to sample.
8. according to the method described in claim 7, wherein, whether include in the image of the identification predetermined danger zone
Before face, the method further includes：
According to reference to Face image synthesis sorter model.
9. according to the method described in claim 8, wherein, whether include people in the image of the identification predetermined danger zone
Face includes：
Extract the feature of described image and generate feature vector；With
Using described eigenvector as the input vector of the sorter model to judge by predetermined danger zone imaging optical path institute
Whether face characteristic is included in the image collected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711437491.6A CN107991283B (en) | 2017-12-26 | 2017-12-26 | Raman spectrum detection device and Raman spectrum detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711437491.6A CN107991283B (en) | 2017-12-26 | 2017-12-26 | Raman spectrum detection device and Raman spectrum detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107991283A true CN107991283A (en) | 2018-05-04 |
| CN107991283B CN107991283B (en) | 2023-09-22 |
Family
ID=62042688
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711437491.6A Active CN107991283B (en) | 2017-12-26 | 2017-12-26 | Raman spectrum detection device and Raman spectrum detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107991283B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104008375A (en) * | 2014-06-04 | 2014-08-27 | 北京工业大学 | Integrated human face recognition mehtod based on feature fusion |
| CN105956515A (en) * | 2016-04-20 | 2016-09-21 | 西安电子科技大学 | Stereo-hyperspectral human face recognition method based on auroral imaging |
| CN106645093A (en) * | 2017-03-21 | 2017-05-10 | 中国工程物理研究院材料研究所 | Raman spectrum plane imaging device |
| CN106770168A (en) * | 2016-12-26 | 2017-05-31 | 同方威视技术股份有限公司 | Article based on Raman spectrum checks device and method |
| CN206292170U (en) * | 2016-12-26 | 2017-06-30 | 同方威视技术股份有限公司 | Article based on Raman spectrum checks equipment |
| CN207779897U (en) * | 2017-12-26 | 2018-08-28 | 同方威视技术股份有限公司 | Raman spectrum detection device |
-
2017
- 2017-12-26 CN CN201711437491.6A patent/CN107991283B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104008375A (en) * | 2014-06-04 | 2014-08-27 | 北京工业大学 | Integrated human face recognition mehtod based on feature fusion |
| CN105956515A (en) * | 2016-04-20 | 2016-09-21 | 西安电子科技大学 | Stereo-hyperspectral human face recognition method based on auroral imaging |
| CN106770168A (en) * | 2016-12-26 | 2017-05-31 | 同方威视技术股份有限公司 | Article based on Raman spectrum checks device and method |
| CN206292170U (en) * | 2016-12-26 | 2017-06-30 | 同方威视技术股份有限公司 | Article based on Raman spectrum checks equipment |
| CN106645093A (en) * | 2017-03-21 | 2017-05-10 | 中国工程物理研究院材料研究所 | Raman spectrum plane imaging device |
| CN207779897U (en) * | 2017-12-26 | 2018-08-28 | 同方威视技术股份有限公司 | Raman spectrum detection device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107991283B (en) | 2023-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Heia et al. | Detection of nematodes in cod (Gadus morhua) fillets by imaging spectroscopy | |
| Yuen et al. | An introduction to hyperspectral imaging and its application for security, surveillance and target acquisition | |
| CN106770168B (en) | Article inspection device and method based on Raman spectrum | |
| CN100415157C (en) | Analysis apparatus and method | |
| CN105912986A (en) | In vivo detection method, in vivo detection system and computer program product | |
| US8253936B2 (en) | Raman characterization of transplant tissue | |
| Chan et al. | The effect of cell fixation on the discrimination of normal and leukemia cells with laser tweezers Raman spectroscopy | |
| JP2019500583A (en) | Computer method and system for deriving spatial proximity between cells | |
| CN111738349B (en) | Detection effect evaluation method and device of target detection algorithm, storage medium and equipment | |
| Alp et al. | Blasts from the Past: Supernova Shock Breakouts among X-Ray Transients in the XMM-Newton Archive | |
| US10921252B2 (en) | Image processing apparatus and method of operating image processing apparatus | |
| CN107860761B (en) | Raman spectrum detection equipment and sample safety detection method for Raman spectrum detection | |
| Mason et al. | Optical technologies for endoscopic real-time histologic assessment of colorectal polyps: a meta-analysis | |
| CN107991287A (en) | Raman spectrum detection device and method based on gradation of image identification | |
| KR101750520B1 (en) | Method for improving detection performance of hyperspectral target detection by using target signal exclusion | |
| US20220214280A1 (en) | Material identification through image capture of raman scattering | |
| CN207779897U (en) | Raman spectrum detection device | |
| JP5938764B2 (en) | DNA double strand breakage analysis apparatus and analysis method | |
| CN107991283A (en) | Raman spectrum detection device and Raman spectra detection process | |
| El Fakir et al. | Active hyperspectral mid-infrared imaging based on a widely tunable quantum cascade laser for early detection of plant water stress | |
| Fischbach et al. | Standoff detection and classification procedure for bioorganic compounds by hyperspectral laser-induced fluorescence | |
| CN108020320A (en) | Raman spectrum detection device and method based on image recognition | |
| WO2015128939A1 (en) | Security evaluation system, security evaluation device, and security evaluation method | |
| CN208488174U (en) | Raman spectrum detection device based on image recognition | |
| CN205620302U (en) | Modular spectrum probe and spectral analysis system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |