CN109658583A - Invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra - Google Patents
Invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra Download PDFInfo
- Publication number
- CN109658583A CN109658583A CN201910104334.6A CN201910104334A CN109658583A CN 109658583 A CN109658583 A CN 109658583A CN 201910104334 A CN201910104334 A CN 201910104334A CN 109658583 A CN109658583 A CN 109658583A
- Authority
- CN
- China
- Prior art keywords
- invoice
- true
- false
- tera
- spectra
- 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
- 238000001228 spectrum Methods 0.000 title claims abstract description 69
- 238000012850 discrimination method Methods 0.000 title claims abstract description 32
- 238000000862 absorption spectrum Methods 0.000 claims abstract description 53
- 238000000605 extraction Methods 0.000 claims abstract description 23
- 238000001328 terahertz time-domain spectroscopy Methods 0.000 claims abstract description 16
- 238000005259 measurement Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims description 11
- 238000002835 absorbance Methods 0.000 claims description 9
- 238000009499 grossing Methods 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 32
- 238000012706 support-vector machine Methods 0.000 description 17
- 230000008569 process Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 239000000126 substance Substances 0.000 description 5
- RTZKZFJDLAIYFH-UHFFFAOYSA-N Diethyl ether Chemical compound CCOCC RTZKZFJDLAIYFH-UHFFFAOYSA-N 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 102000008297 Nuclear Matrix-Associated Proteins Human genes 0.000 description 1
- 108010035916 Nuclear Matrix-Associated Proteins Proteins 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 210000000299 nuclear matrix Anatomy 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/06—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using wave or particle radiation
- G07D7/12—Visible light, infrared or ultraviolet radiation
- G07D7/128—Viewing devices
Landscapes
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Toxicology (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
This application discloses a kind of invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra, this method comprises: being pre-processed to true invoice and false invoice；True invoice and false invoice are respectively placed in the measured zone of terahertz time-domain spectroscopy instrument and are taken multiple measurements, generates terahertz absorption spectra figure according to collected data are measured；Specific section is selected to carry out feature extraction in terahertz absorption spectra figure；It is trained by feature of the SVM classifier to extraction, obtains the cut-off rule for identifying true invoice and false invoice.The application obtains identifying the cut-off rule of true invoice and false invoice using tera-hertz spectra and SVM classifier, the difference of true invoice and false invoice can be obviously distinguished by the cut-off rule, and then can quickly identify true and false invoice, the accuracy that invoice identifies is improved, the error rate for judging true and false invoice is reduced.
Description
Technical field
The present invention relates to substance authentication technique fields, more particularly to a kind of invoice identification side based on tera-hertz spectra
Method, device, equipment and storage medium.
Background technique
Invoice refers to all entity and individual in purchasing and selling commodities and is engaged in the business that other business activities issue and collect
Voucher is the important evidence of accounting, auditing authority, tax authority's examination of law enforcement, common invoice and increasing is divided into from management
Two kinds of tax invoice of value.
In recent years, invoice occurs faking and more and more savage, and some criminals attempt to obtain using false invoice more
Interests, false invoice is compared with true invoice, hence it is evident that difference is invoice papery, Invoice and affixes one's seal.And with the hair of science and technology
Exhibition, present forgery technology is more and more true to nature, is badly in need of a kind of effective recognition methods to true and false invoice.It is true for identifying at present
The method of false invoice is generally ultraviolet light differential method, deficiency be exactly identify invoice accuracy it is not high.
Therefore, the accuracy that true and false invoice identifies how is improved, is those skilled in the art's technical problem urgently to be resolved.
Summary of the invention
In view of this, the invoice discrimination method that the purpose of the present invention is to provide a kind of based on tera-hertz spectra, device, setting
Standby and storage medium can quickly identify true and false invoice, improve the accuracy that invoice identifies, and reduce the mistake for judging true and false invoice
Rate.Its concrete scheme is as follows:
A kind of invoice discrimination method based on tera-hertz spectra, comprising:
True invoice and false invoice are pre-processed；
The true invoice and the false invoice are respectively placed in the measured zone of terahertz time-domain spectroscopy instrument and carried out repeatedly
Measurement generates terahertz absorption spectra figure according to collected data are measured；
Specific section is selected to carry out feature extraction in the terahertz absorption spectra figure；
It is trained by feature of the SVM classifier to extraction, obtains point for identifying the true invoice and the false invoice
Secant.
Preferably, in the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention, in basis
After measuring collected data generation terahertz absorption spectra figure, specific area is selected in the terahertz absorption spectra figure
Between carry out feature extraction before, further includes:
The terahertz absorption spectra figure is smoothed using gaussian filtering exponential smoothing.
Preferably, in the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention, according to survey
It measures collected data and generates terahertz absorption spectra figure, specifically include:
According to collected data are measured, the terahertz time-domain spectroscopy of the true invoice and the false invoice is obtained；
The terahertz time-domain spectroscopy that will acquire carries out Fourier transformation, is converted to power spectrum；
According to the power spectrum being converted to, transmissivity is calculated；
According to the calculated transmissivity, absorbance is calculated；
According to the calculated absorbance, the terahertz absorption spectra figure of the true invoice and the false invoice is obtained.
Preferably, it in the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention, is obtaining
After the cut-off rule for identifying the true invoice and the false invoice, further includes:
It is using the obtained cut-off rule as boundary, the terahertz absorption spectra figure below the cut-off rule is corresponding
Invoice be determined as true invoice；The corresponding invoice of the terahertz absorption spectra figure above the cut-off rule is determined as vacation
Invoice.
The embodiment of the invention also provides a kind of invoice identification device based on tera-hertz spectra, comprising:
Invoice preprocessing module, for being pre-processed to true invoice and false invoice；
Spectrogram generation module, for the true invoice and the false invoice to be respectively placed in terahertz time-domain spectroscopy instrument
It is taken multiple measurements in measured zone, generates terahertz absorption spectra figure according to collected data are measured；
Characteristic extracting module, for selecting specific section to carry out feature extraction in the terahertz absorption spectra figure；
SVM training module obtains identifying the true invoice for being trained by feature of the SVM classifier to extraction
With the cut-off rule of the false invoice.
Preferably, in the above-mentioned invoice identification device based on tera-hertz spectra provided in an embodiment of the present invention, further includes:
Spectrogram Leveling Block, for smoothly being located using gaussian filtering exponential smoothing to the terahertz absorption spectra figure
Reason.
Preferably, in the above-mentioned invoice identification device based on tera-hertz spectra provided in an embodiment of the present invention, further includes:
Invoice determination module, for using the obtained cut-off rule as boundary, by described in below the cut-off rule too
The corresponding invoice of hertz abosrption spectrogram is determined as true invoice；By the terahertz absorption spectra figure above the cut-off rule
Corresponding invoice is determined as false invoice.
The embodiment of the invention also provides a kind of invoice authentication equipment based on tera-hertz spectra, including processor and storage
Device, wherein the processor is realized when executing the computer program saved in the memory as provided in an embodiment of the present invention
The above-mentioned invoice discrimination method based on tera-hertz spectra.
The embodiment of the invention also provides a kind of computer readable storage mediums, for storing computer program, wherein institute
It states and is realized when computer program is executed by processor such as the above-mentioned invoice mirror based on tera-hertz spectra provided in an embodiment of the present invention
Other method.
A kind of invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra provided by the present invention, should
Method includes: to pre-process to true invoice and false invoice；True invoice and false invoice are respectively placed in terahertz time-domain spectroscopy instrument
Measured zone in take multiple measurements, generate terahertz absorption spectra figure according to collected data are measured；It is inhaled in Terahertz
It receives and specific section is selected to carry out feature extraction in spectrogram；It is trained, is reflected by feature of the SVM classifier to extraction
The cut-off rule of untrue invoice and false invoice.
The present invention obtains identifying the cut-off rule of true invoice and false invoice using tera-hertz spectra and SVM classifier, by this
Cut-off rule can obviously distinguish the difference of true invoice and false invoice, and then can quickly identify true and false invoice, improve invoice mirror
Other accuracy reduces the error rate for judging true and false invoice.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is the flow chart of the invoice discrimination method provided in an embodiment of the present invention based on tera-hertz spectra；
Fig. 2 is the specific flow chart of the invoice discrimination method provided in an embodiment of the present invention based on tera-hertz spectra；
Fig. 3 is the terahertz absorption spectra figure of the false invoice provided in an embodiment of the present invention without smoothing processing；
Fig. 4 is the terahertz absorption spectra figure of the true invoice provided in an embodiment of the present invention without smoothing processing；
Fig. 5 is the terahertz absorption spectra figure of true and false invoice after smoothed processing provided in an embodiment of the present invention；
Fig. 6 is the terahertz of the true and false invoice comprising cut-off rule after the training provided in an embodiment of the present invention using SVM classifier
Hereby abosrption spectrogram；
Fig. 7 is the structural schematic diagram of the invoice identification device provided in an embodiment of the present invention based on tera-hertz spectra.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention provides a kind of invoice discrimination method based on tera-hertz spectra, as shown in Figure 1, comprising the following steps:
S101, true invoice and false invoice are pre-processed；
S102, true invoice and false invoice are respectively placed in the measured zone of terahertz time-domain spectroscopy instrument and repeatedly surveyed
Amount generates terahertz absorption spectra figure according to collected data are measured；
S103, specific section is selected to carry out feature extraction in terahertz absorption spectra figure；
S104, it is trained by feature of the SVM classifier to extraction, obtains the segmentation for identifying true invoice and false invoice
Line.
In the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention, first to true invoice and
False invoice is pre-processed；Then true invoice and false invoice are respectively placed in the measured zone of terahertz time-domain spectroscopy instrument and are carried out
Repeatedly measurement generates terahertz absorption spectra figure according to collected data are measured；It is selected in terahertz absorption spectra figure later
It selects specific section and carries out feature extraction；It is trained finally by feature of the SVM classifier to extraction, obtains identifying true invoice
With the cut-off rule of false invoice.Obtain identifying the cut-off rule of true invoice and false invoice using tera-hertz spectra and SVM classifier in this way,
The difference of true invoice and false invoice can be obviously distinguished by the cut-off rule, and then can quickly identify true and false invoice, improved
The accuracy that invoice identifies reduces the error rate for judging true and false invoice.
It should be noted that THz wave is between visible light and microwave, audio range frequency is often referred to 0.1THz extremely
Between 10THz, belong to far infrared band.Since THz wave has the properties such as stronger penetrability, fingerprint spectrality, low energy,
It meets substance and identifies most concerned factor, substance can be identified by analyzing the feature of Terahertz spectrum.Therefore, this hair
It is bright that invoice is identified using THz wave, extremely strong feature is penetrated to apolar substance using it, obtains transmitted spectrum information, together
Shi Buyi damages the invoice being authenticated.
Further, in the specific implementation, in the above-mentioned invoice mirror based on tera-hertz spectra provided in an embodiment of the present invention
In other method, for the ease of the feature of the true invoice of observation and the terahertz absorption spectra figure of false invoice, as shown in Fig. 2, executing
After step S102 is according to collected data generation terahertz absorption spectra figure is measured, inhaled executing step S103 in Terahertz
Receive before selecting specific section to carry out feature extraction in spectrogram, can with the following steps are included:
S201, terahertz absorption spectra figure is smoothed using gaussian filtering exponential smoothing.
Specifically, elimination noise is carried out to the terahertz absorption spectra figure of invoice, smoothed curve is smooth using gaussian filtering
Method, first reference Gaussian function:
Discretization is carried out to Gaussian function, using the Gaussian function numerical value on discrete point as weight, to collected gray matrix
Each pixel do a certain range of weighted average, Gaussian noise can be eliminated.Discrete Gaussian convolution core, element calculate
Method are as follows:
Wherein, σ is variance, and k is the dimension of nuclear matrix.
Gaussian template is found out with can be convenient by above-mentioned formula (2), to carry out gaussian filtering process.
In the specific implementation, in the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention,
Step S101 pre-processes true invoice and false invoice, can specifically include: collecting true invoice and false invoice first as instruction
Practice sample；Then in setting humidity range, the true invoice and false invoice of collection are flattened on press.To both
Invoice is flattened on press, is that invoice has fold in order to prevent, press is suitable for the thin paper of different-thickness, Ke Yiyun
It uses on invoice, it is flattened, make the smooth surface of invoice, reduction causes a deviation to training result.It is being pre-processed
During, the humidity of environment to be measured can be measured with air humidity instrument, humidity environment is necessary for 10% hereinafter, guaranteeing result
Accuracy.
In the specific implementation, in the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention,
Terahertz time-domain spectroscopy instrument used in step S102 can select TAS7400 serial, can measure 0.1THz~5THz too
Hertz spectrum, is capable of providing many kinds of substance parameter such as power, phase, transmitance, reflectivity, refractive index, dielectric constant, using too
The characteristic of Hertz wave, is used for nondestructive spectrum analysis, and fully automatic system retains data.
Step S102 generates terahertz absorption spectra figure according to collected data are measured, and can specifically include:
The first step, according to collected data are measured, obtain the terahertz time-domain spectroscopy of true invoice and false invoice；
Second step, the terahertz time-domain spectroscopy that will acquire carry out Fourier transformation, are converted to power spectrum；
Power spectrum is converted according to following equation (3):
Wherein, PLinear(ω) is power spectrum,It is FFT data (plural number)；
It is absolute value of a complex number.
The power spectrum that third step, basis are converted to, calculates transmissivity；
Transmissivity is calculated according to following equation (4):
Wherein TLinear(ω) is transmissivity, Psam(ω) is sample power spectrum, Pref(ω) is reference power spectrum.
4th step, according to calculated transmissivity, calculate absorbance；
Absorbance is according to following calculation formula:
Wherein A (ω) represents absorbance, TLinear(ω) is transmissivity.
5th step, according to calculated absorbance, obtain the terahertz absorption spectra figure of true invoice and false invoice；
Specifically, ether hertz frequency is abscissa, and absorbance is ordinate, as shown in Figure 3 and Figure 4, is drawn out respectively
The terahertz absorption spectra figure of false invoice and true invoice.The terahertz absorption spectra figure of Fig. 3 and Fig. 4 is merged and carried out
After smoothing processing, available spectrogram as shown in Figure 5.
In the specific implementation, in the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention,
During executing step S103, it is assumed that curve set sample S={ (x1,y1)...(xn,yn), select specific section are as follows:
Q=(c, d) | c=xa, d=xb}
This it appears that the peak value in this section of section 1.5THz to 3THz is all obvious from Fig. 5, i.e., selected xa=
1.5,xb=3 convenient for the true and false invoice of differentiation.Also, in specific section, the difference of two curves can be obviously observed in Fig. 5
Off course degree, the peak value of false invoice are apparently higher than the peak value of true invoice.
In the specific implementation, in the above-mentioned invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention,
SVM (support vector machines) principle is that commonplace one kind in Statistical Learning Theory and artificial intelligence is logical in step S104
Learning method, it can be trained the study essence i.e. to sample training according to the complexity of sample to specified sample
Degree makes accurate judgement to expected sample by trained sample data.
For Linear SVM as a kind of two classifiers, core is exactly to find a hyperplane, and this hyperplane is from all points
All as far as possible far, this hyperplane is exactly straight line, corresponding equation are as follows: y=wTX+b, wherein wTFor normal vector, b
For intercept.For its corresponding equation of two-dimensional surface:
(w1,w2) (x, y)+b=0 → w1·x+w2Y+b=0 (6)
The range formula of straight line is arrived according to point, the distance of each training sample S to straight line:
Wherein, i=1,2,3 ..., the range formula for having training sample S to straight line just conveniently obtains geometry interval:
The classification of true and false invoice is distinguished, the principle of classification is to find a hyperplane for the curve and false invoice of true invoice
Curve it is separated as far as possible, then a maximum spacing should be found at this time:
Wherein s.t is restrictive condition.
Above formula (9) are solved, a convex quadratic programming problem is obtained, it, can be with since the scaling of the solution of quadratic programming does not influence
Simplify and calculates separatelyIt is then converted into linear separability and supports its expression of training problem:
With cut-off rule wT·xi+ b=1 is boundary, thus the comparison to judge true and false invoice He this formula.As shown in fig. 6,
The straight line being directed toward in spectrogram with arrow is exactly the cut-off rule of identification true invoice and false invoice that the present invention obtains.
Further, in the specific implementation, in the above-mentioned invoice mirror based on tera-hertz spectra provided in an embodiment of the present invention
In other method, as shown in Fig. 2, can also be wrapped after executing step S104 and obtaining identifying the cut-off rule of true invoice and false invoice
Include following steps:
S202, using obtained cut-off rule as boundary, by the corresponding invoice of terahertz absorption spectra figure below cut-off rule
It is determined as true invoice；The corresponding invoice of terahertz absorption spectra figure above cut-off rule is determined as false invoice.
Specifically, work as wT·xi+ b≤1, i.e. training standard line thereunder, can judge that it is true invoice；Work as wT·
xi+ b >=1, i.e. training standard line are above it, it can be determined that it is false invoice.Input the data of sample to be tested, so that it may treat
Test sample originally carries out the differentiation of true and false invoice.
Based on the same inventive concept, the embodiment of the invention also provides a kind of, and the invoice based on tera-hertz spectra identifies dress
It sets, the principle that is solved the problems, such as due to the invoice identification device based on tera-hertz spectra and aforementioned a kind of based on tera-hertz spectra
Invoice discrimination method is similar, therefore the implementation for being somebody's turn to do the invoice identification device based on tera-hertz spectra may refer to based on terahertz light
The implementation of the invoice discrimination method of spectrum, overlaps will not be repeated.
In the specific implementation, the invoice identification device provided in an embodiment of the present invention based on tera-hertz spectra, such as Fig. 7 institute
Show, specifically include:
Invoice preprocessing module 11, for being pre-processed to true invoice and false invoice；
Spectrogram generation module 12, for true invoice and false invoice to be respectively placed in the measurement zone of terahertz time-domain spectroscopy instrument
It is taken multiple measurements in domain, generates terahertz absorption spectra figure according to collected data are measured；
Characteristic extracting module 13, for selecting specific section to carry out feature extraction in terahertz absorption spectra figure；
SVM training module 14, for being trained by feature of the SVM classifier to extraction, obtain identifying true invoice and
The cut-off rule of false invoice.
In the above-mentioned invoice identification device based on tera-hertz spectra provided in an embodiment of the present invention, above-mentioned four can be passed through
The interaction of a module, the cut-off rule obtained using tera-hertz spectra and SVM classifier obviously distinguish true invoice and wig
The difference of ticket, and then can quickly identify true and false invoice, the accuracy that invoice identifies is improved, the mistake for judging true and false invoice is reduced
Rate.
Further, in the specific implementation, in the above-mentioned invoice mirror based on tera-hertz spectra provided in an embodiment of the present invention
In other device, for the ease of observe true invoice and false invoice terahertz absorption spectra figure feature, as shown in fig. 7, can be with
Include:
Spectrogram Leveling Block 15, it is smooth for being carried out using gaussian filtering exponential smoothing to the terahertz absorption spectra figure
Processing.
Further, in the specific implementation, in the above-mentioned invoice based on tera-hertz spectra provided in an embodiment of the present invention
In identification device, as shown in fig. 7, can also include:
Invoice determination module 16, for using obtained cut-off rule as boundary, the Terahertz below cut-off rule to be absorbed light
The corresponding invoice of spectrogram is determined as true invoice；The corresponding invoice of terahertz absorption spectra figure above cut-off rule is determined as vacation
Invoice.
Corresponding contents disclosed in previous embodiment can be referred to about the more specifical course of work of above-mentioned modules,
This is no longer repeated.
Correspondingly, the embodiment of the invention also discloses a kind of invoice authentication equipment based on tera-hertz spectra, including processing
Device and memory；Wherein, it realizes when processor executes the computer program saved in memory and is based on disclosed in previous embodiment
The invoice discrimination method of tera-hertz spectra.
It can be with reference to corresponding contents disclosed in previous embodiment, herein no longer about the more specifical process of the above method
It is repeated.
Further, the invention also discloses a kind of computer readable storage mediums, for storing computer program；It calculates
Machine program realizes the aforementioned disclosed invoice discrimination method based on tera-hertz spectra when being executed by processor.
It can be with reference to corresponding contents disclosed in previous embodiment, herein no longer about the more specifical process of the above method
It is repeated.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
It sets, for equipment, storage medium, since it is corresponded to the methods disclosed in the examples, so be described relatively simple, correlation
Place is referring to method part illustration.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond scope of the present application.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
To sum up, it a kind of invoice discrimination method based on tera-hertz spectra provided in an embodiment of the present invention, device, equipment and deposits
Storage media, this method comprises: being pre-processed to true invoice and false invoice；When true invoice and false invoice are respectively placed in Terahertz
It is taken multiple measurements in the measured zone of domain spectrometer, generates terahertz absorption spectra figure according to collected data are measured；?
Specific section is selected to carry out feature extraction in terahertz absorption spectra figure；It is instructed by feature of the SVM classifier to extraction
Practice, obtains the cut-off rule for identifying true invoice and false invoice.It obtains identifying true invoice using tera-hertz spectra and SVM classifier in this way
With the cut-off rule of false invoice, the difference of true invoice and false invoice can be obviously distinguished by the cut-off rule, and then can be quick
Identify true and false invoice, improve the accuracy that invoice identifies, reduces the error rate for judging true and false invoice.
Finally, it is to be noted that, herein, relational terms be used merely to by an entity or operation with it is another
A entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this actual
Relationship or sequence.Moreover, the terms "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion,
So that the process, method, article or equipment for including a series of elements not only includes those elements, but also including not having
The other element being expressly recited, or further include for elements inherent to such a process, method, article, or device.Do not having
There is the element limited in the case where more limiting by sentence "including a ...", it is not excluded that in the mistake including the element
There is also other identical elements in journey, method, article or equipment.
Invoice discrimination method, device, equipment and storage medium to provided by the present invention based on tera-hertz spectra above
It is described in detail, used herein a specific example illustrates the principle and implementation of the invention, the above reality
The explanation for applying example is merely used to help understand method and its core concept of the invention；Meanwhile for the general technology of this field
Personnel, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion this theory
Bright book content should not be construed as limiting the invention.
Claims (9)
1. a kind of invoice discrimination method based on tera-hertz spectra characterized by comprising
True invoice and false invoice are pre-processed；
The true invoice and the false invoice are respectively placed in the measured zone of terahertz time-domain spectroscopy instrument and taken multiple measurements,
Terahertz absorption spectra figure is generated according to collected data are measured；
Specific section is selected to carry out feature extraction in the terahertz absorption spectra figure；
It is trained by feature of the SVM classifier to extraction, obtains the cut-off rule for identifying the true invoice and the false invoice.
2. the invoice discrimination method according to claim 1 based on tera-hertz spectra, which is characterized in that adopted according to measurement
After the data collected generate terahertz absorption spectra figure, specific section is selected to carry out in the terahertz absorption spectra figure
Before feature extraction, further includes:
The terahertz absorption spectra figure is smoothed using gaussian filtering exponential smoothing.
3. the invoice discrimination method according to claim 1 based on tera-hertz spectra, which is characterized in that acquired according to measurement
The data arrived generate terahertz absorption spectra figure, specifically include:
According to collected data are measured, the terahertz time-domain spectroscopy of the true invoice and the false invoice is obtained；
The terahertz time-domain spectroscopy that will acquire carries out Fourier transformation, is converted to power spectrum；
According to the power spectrum being converted to, transmissivity is calculated；
According to the calculated transmissivity, absorbance is calculated；
According to the calculated absorbance, the terahertz absorption spectra figure of the true invoice and the false invoice is obtained.
4. the invoice discrimination method according to claim 2 based on tera-hertz spectra, which is characterized in that obtaining identifying institute
After the cut-off rule for stating true invoice and the false invoice, further includes:
Using the obtained cut-off rule as boundary, by the corresponding hair of the terahertz absorption spectra figure below the cut-off rule
Ticket is determined as true invoice；The corresponding invoice of the terahertz absorption spectra figure above the cut-off rule is determined as wig
Ticket.
5. a kind of invoice identification device based on tera-hertz spectra characterized by comprising
Invoice preprocessing module, for being pre-processed to true invoice and false invoice；
Spectrogram generation module, for the true invoice and the false invoice to be respectively placed in the measurement of terahertz time-domain spectroscopy instrument
It is taken multiple measurements in region, generates terahertz absorption spectra figure according to collected data are measured；
Characteristic extracting module, for selecting specific section to carry out feature extraction in the terahertz absorption spectra figure；
SVM training module obtains identifying the true invoice and institute for being trained by feature of the SVM classifier to extraction
State the cut-off rule of false invoice.
6. the invoice discrimination method according to claim 5 based on tera-hertz spectra, which is characterized in that further include:
Spectrogram Leveling Block, for being smoothed using gaussian filtering exponential smoothing to the terahertz absorption spectra figure.
7. the invoice discrimination method according to claim 6 based on tera-hertz spectra, which is characterized in that further include:
Invoice determination module, for using the obtained cut-off rule as boundary, by the Terahertz below the cut-off rule
The corresponding invoice of abosrption spectrogram is determined as true invoice；The terahertz absorption spectra figure above the cut-off rule is corresponding
Invoice be determined as false invoice.
8. a kind of invoice authentication equipment based on tera-hertz spectra, which is characterized in that including processor and memory, wherein institute
Realization when processor executes the computer program saved in the memory is stated to be based on as Claims 1-4 is described in any item
The invoice discrimination method of tera-hertz spectra.
9. a kind of computer readable storage medium, which is characterized in that for storing computer program, wherein the computer journey
Such as Claims 1-4 described in any item invoice discrimination methods based on tera-hertz spectra are realized when sequence is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910104334.6A CN109658583A (en) | 2019-02-01 | 2019-02-01 | Invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910104334.6A CN109658583A (en) | 2019-02-01 | 2019-02-01 | Invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109658583A true CN109658583A (en) | 2019-04-19 |
Family
ID=66122609
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910104334.6A Pending CN109658583A (en) | 2019-02-01 | 2019-02-01 | Invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109658583A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1675657A (en) * | 2002-06-19 | 2005-09-28 | 马尔斯公司 | Currency validator |
| US20080036999A1 (en) * | 2004-12-18 | 2008-02-14 | Wolfgang Petrich | Spectroscopic examination of body fluid and tissue samples to aid diagnosis of alzheimer's disease |
| CN101532953A (en) * | 2009-04-02 | 2009-09-16 | 中国计量学院 | Method for accurately measuring optical parameters of edible oil |
| JP2011034173A (en) * | 2009-07-30 | 2011-02-17 | Glory Ltd | Method and apparatus for determining authenticity of paper sheet using terahertz light |
| CN102147941A (en) * | 2011-02-28 | 2011-08-10 | 浙江大学 | Renminbi selectable multiple-characteristic-point transmission authenticity-identifying method based on terahertz time-domain spectroscopy |
| CN102175636A (en) * | 2011-03-18 | 2011-09-07 | 上海理工大学 | System and method for detecting and identifying Chinese herbal medicine |
| CN103714616A (en) * | 2012-10-04 | 2014-04-09 | 起山电子株式会社 | Method and device of discriminating a banknote by using terahertz electromagnetic waves |
| CN103822884A (en) * | 2013-12-19 | 2014-05-28 | 中国矿业大学 | Classified identification method for tea based on terahertz time-domain spectroscopy |
| CN102590135B (en) * | 2012-03-02 | 2014-06-04 | 中国计量学院 | Herbicide distinguishing method based on least-square support vector machine |
| CN104406923A (en) * | 2014-11-14 | 2015-03-11 | 首都师范大学 | Method for quantitative detection of amino acid content in grain by THz-TDS technology |
| CN107331032A (en) * | 2017-07-11 | 2017-11-07 | 中央民族大学 | A kind of method of the identification note true and false |
| CN108760652A (en) * | 2018-03-26 | 2018-11-06 | 南京农业大学 | The more quality damage-free detection methods of grape based on Vis/NIR and device |
-
2019
- 2019-02-01 CN CN201910104334.6A patent/CN109658583A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1675657A (en) * | 2002-06-19 | 2005-09-28 | 马尔斯公司 | Currency validator |
| US20080036999A1 (en) * | 2004-12-18 | 2008-02-14 | Wolfgang Petrich | Spectroscopic examination of body fluid and tissue samples to aid diagnosis of alzheimer's disease |
| CN101532953A (en) * | 2009-04-02 | 2009-09-16 | 中国计量学院 | Method for accurately measuring optical parameters of edible oil |
| JP2011034173A (en) * | 2009-07-30 | 2011-02-17 | Glory Ltd | Method and apparatus for determining authenticity of paper sheet using terahertz light |
| CN102147941A (en) * | 2011-02-28 | 2011-08-10 | 浙江大学 | Renminbi selectable multiple-characteristic-point transmission authenticity-identifying method based on terahertz time-domain spectroscopy |
| CN102175636A (en) * | 2011-03-18 | 2011-09-07 | 上海理工大学 | System and method for detecting and identifying Chinese herbal medicine |
| CN102590135B (en) * | 2012-03-02 | 2014-06-04 | 中国计量学院 | Herbicide distinguishing method based on least-square support vector machine |
| CN103714616A (en) * | 2012-10-04 | 2014-04-09 | 起山电子株式会社 | Method and device of discriminating a banknote by using terahertz electromagnetic waves |
| CN103822884A (en) * | 2013-12-19 | 2014-05-28 | 中国矿业大学 | Classified identification method for tea based on terahertz time-domain spectroscopy |
| CN104406923A (en) * | 2014-11-14 | 2015-03-11 | 首都师范大学 | Method for quantitative detection of amino acid content in grain by THz-TDS technology |
| CN107331032A (en) * | 2017-07-11 | 2017-11-07 | 中央民族大学 | A kind of method of the identification note true and false |
| CN108760652A (en) * | 2018-03-26 | 2018-11-06 | 南京农业大学 | The more quality damage-free detection methods of grape based on Vis/NIR and device |
Non-Patent Citations (7)
| Title |
|---|
| CHEN XIAI,ET AL: "Classification of the green tea varieties based on Support Vector Machines using Terahertz spectroscopy", 《INSTRUMENTATION & MEASUREMENT TECHNOLOGY CONFERENCE》 * |
| LIU J,ET AL: "Organic Compound Identification Based on Terahertz Spectrum", 《2018 IEEE 7TH DATA DRIVEN CONTROL AND LEARNING SYSTEMS CONFERENCE (DDCLS)》 * |
| 刘海玲等: "基于THz光谱成像的纸币真伪鉴别研究", 《光谱学与光谱分析》 * |
| 康琦等: "《机器学习中的不平衡分类方法》", 31 October 2017, 同济大学出版社 * |
| 李晗等: "基于时域太赫兹波谱法的降糖类药物分类与分析", 《医用生物力学》 * |
| 李武等: "基于主成分分析和支持向量机的太赫兹光谱冰片鉴别", 《光谱学与光谱分析》 * |
| 杨琴等: "基于激光光致发光光谱的发票真伪鉴别方法研究", 《光谱学与光谱分析》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rodenacker et al. | A feature set for cytometry on digitized microscopic images | |
| JP6366556B2 (en) | Method for analyzing biological specimens with spectral images | |
| CN103528617B (en) | A kind of cockpit instrument identifies and detection method and device automatically | |
| JP6000345B2 (en) | Analysis method of biological samples by spectral imaging | |
| Heer et al. | Multi-scale banking to 45 degrees | |
| CN101498661B (en) | Infrared spectrum characteristic extraction method for high-precision discriminating variety, producing area and growth mode of traditional Chinese medicinal materials | |
| CN110068544A (en) | Material identification network model training method and tera-hertz spectra substance identification | |
| CN107818298A (en) | General Raman spectral characteristics extracting method for machine learning material recognition | |
| CN106198482A (en) | The method whether being added with Western medicine in detection health product based on Raman spectrum | |
| CN109360089A (en) | Credit risk prediction technique and device | |
| CN107679569A (en) | Raman spectrum substance automatic identifying method based on adaptive hypergraph algorithm | |
| CN109146003B (en) | Hyperspectral imaging-based classification identification method, device and system | |
| CN111523587A (en) | Woody plant species spectrum identification method based on machine learning | |
| Zhang et al. | Spectral and chromatographic overall analysis: An insight into chemical equivalence assessment of traditional Chinese medicine | |
| Wang et al. | End‐to‐end analysis modeling of vibrational spectroscopy based on deep learning approach | |
| Peel et al. | Assessing the performance of rational spline-based empirical mode decomposition using a global annual precipitation dataset | |
| CN112232343B (en) | Grain mildew grain identification neural network and identification method | |
| US20210199643A1 (en) | Fluid classification | |
| CN109658583A (en) | Invoice discrimination method, device, equipment and storage medium based on tera-hertz spectra | |
| Baker et al. | Analysis of tools used to quantify droplet clustering in clouds | |
| CN110987856A (en) | Cosmetic quality rapid identification method based on formula system and fingerprint spectrum | |
| Lang et al. | Breast cancer magnification-independent multi-class histopathology classification using dual-step model | |
| CN114821618A (en) | Analysis method for OFD reading software display effect | |
| JPH03179281A (en) | Analyzing method for spectrum of signal and displaying method for result of analysis | |
| Polejowska et al. | Impact of Visual Image Quality on Lymphocyte Detection Using YOLOv5 and RetinaNet Algorithms |
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: 20190419 |
|
| RJ01 | Rejection of invention patent application after publication |