CN106910276B - Detect the new and old method and device of bank note - Google Patents
Detect the new and old method and device of bank note Download PDFInfo
- Publication number
- CN106910276B CN106910276B CN201710103550.XA CN201710103550A CN106910276B CN 106910276 B CN106910276 B CN 106910276B CN 201710103550 A CN201710103550 A CN 201710103550A CN 106910276 B CN106910276 B CN 106910276B
- Authority
- CN
- China
- Prior art keywords
- value
- bank note
- new
- detected
- sobel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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
- G07D7/20—Testing patterns thereon
- G07D7/2016—Testing patterns thereon using feature extraction, e.g. segmentation, edge detection or Hough-transformation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention discloses a kind of new and old method and devices of detection bank note, the described method includes: being based on Sobel Operator template, the grey scale pixel value at least one fisrt feature region in bank note to be detected is converted, the vertical features value and horizontal properties value at least one fisrt feature region are obtained；According to the vertical features value, the horizontal properties value and predetermined level equation group, determine the new and old grade of the bank note to be detected, the equation quantity for wherein including in the predetermined level equation group is equal with the total quantity of the vertical features value and the horizontal properties value.The embodiment of the present invention solves the problems, such as that the new and old recognition effect of bank note is poor in the prior art, improves the accuracy of the new and old identification of bank note.
Description
Technical field
The present embodiments relate to bank note inspection technology field more particularly to the methods and dress of a kind of new and old detection of bank note
It sets.
Background technique
Requirement due to different field to circulating paper money quality is different, it is therefore necessary to examine to the new and old of bank note
It surveys.
Currently, the new and old method of inspection of bank note has image detection, Thickness sensitivity, sound detection, chemical detection etc..Wherein, scheme
It is the most commonly used as detecting.For example, the gray value mean value of statistics banknote image partial region, by the standard gray angle value of itself and the region
It is matched, the new and old grade of the bank note can be obtained.
But the precision of the new aging method of image detection bank note in the prior art is low, the new and old recognition effect of bank note is poor.
Summary of the invention
The embodiment of the present invention provides a kind of method and device of new and old detection of bank note, can effectively improve the new and old grade of bank note
The accuracy of detection.
In a first aspect, the embodiment of the invention provides a kind of methods of the new and old detection of bank note, comprising:
Based on Sobel Operator template, the grey scale pixel value at least one fisrt feature region in bank note to be detected is carried out
Conversion obtains the vertical features value and horizontal properties value at least one fisrt feature region；
According to the vertical features value, the horizontal properties value and predetermined level equation group, the paper to be detected is determined
The new and old grade of coin, wherein the equation quantity for including in the predetermined level equation group, with the vertical features value and the water
The total quantity of flat characteristic value is equal.
Second aspect, the embodiment of the invention also provides a kind of devices of the new and old detection of bank note, comprising:
Characteristic value acquisition module, for being based on Sobel Operator template, at least one fisrt feature in bank note to be detected
The grey scale pixel value in region is converted, and the vertical features value and horizontal properties at least one fisrt feature region are obtained
Value；
The first estate determining module, for according to the vertical features value, the horizontal properties value and predetermined level side
Journey group determines the new and old grade of the bank note to be detected, wherein the equation quantity for including in the predetermined level equation group, with institute
The total quantity for stating vertical features value and the horizontal properties value is equal.
The embodiment of the invention provides a kind of method and devices of the new and old detection of bank note, are obtained based on Sobel Operator template
Vertical features value and horizontal properties value determine the new and old grade of bank note to be detected in conjunction with predetermined level equation group, solve existing
There is the problem that the new and old recognition effect of bank note is poor in technology, improves the accuracy of the new and old identification of bank note.
Detailed description of the invention
Fig. 1 is the flow chart of the new and old detection method of one of the embodiment of the present invention one bank note；
Fig. 2A is the flow chart of the new and old detection method of one of the embodiment of the present invention two bank note；
Fig. 2 B is the selected schematic diagram in fisrt feature region and second feature region in the embodiment of the present invention two；
Fig. 3 is the flow chart of the new and old detection method of one of the embodiment of the present invention three bank note；
Fig. 4 is the flow chart of the new and old detection method of one of the embodiment of the present invention four bank note；
Fig. 5 is the structure chart of the new and old detection device of one of the embodiment of the present invention five bank note.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is a kind of flow chart for the new and old detection method of bank note that the embodiment of the present invention one provides, and the present embodiment is applicable
In the new and old detection of various bank note the case where, this method can be held by the device of the new and old detection of bank note provided in an embodiment of the present invention
Row, the mode which can be used software and/or hardware realize that the device can be integrated in the new and old detection function of any offer bank note
Equipment in, such as equipment can be ATM (Automatic Teller Machine, ATM), paper currency sorter, such as
Shown in Fig. 1, specifically include:
S110, it is based on Sobel Operator template, to the pixel grey scale at least one fisrt feature region in bank note to be detected
Value is converted, and the vertical features value and horizontal properties value at least one fisrt feature region are obtained.
Wherein, Sobel Operator (Sobel operator) template is one of the operator template in image procossing, includes two
The matrix of group 3 × 3, respectively transverse direction and longitudinal direction, make planar convolution with image for it, can obtain the bright of transverse direction and longitudinal direction respectively
Spend difference approximation value.Sobel Operator template includes Sobel Operator vertical formwork and Sobel Operator horizontal shuttering, main use
Make detection vertical edge and horizontal edge.Bank note to be detected can obtain grey scale pixel value, grey scale pixel value by Image Acquisition
For the gray value of pixel each in image, wherein gray scale be black white image midpoint color depth, range generally from 0 to
255.The region that gray value is relatively uniform in bank note is chosen in fisrt feature region, can choose the front of bank note, can also choose
The reverse side of bank note.Also, fisrt feature region can be a characteristic area, or multiple characteristic areas, area are big
It is small to be not construed as limiting.Vertical features value is after each pixel in fisrt feature region is converted according to Sobel vertical formwork
Conversion value, horizontal properties value be fisrt feature region in each pixel converted according to Sobel horizontal shuttering after
Conversion value.Likewise, vertical features value and horizontal properties value may be one or more, number with selection first
Characteristic area number is identical.
Optionally, the region that grey scale pixel value in the bank note to be detected is greater than pixel threshold value is determined as described to be checked
Survey the characteristic area of bank note.
Specifically, the fisrt feature region of bank note to be detected can be extracted according to grey scale pixel value.Grey scale pixel value
Higher, display area is whiter.During routine use bank note, due in bank note the lower region of grey scale pixel value by new and old shadow
Sound is smaller, thus identify bank note to be detected it is new and old in often the higher region of selected pixels gray value as area to be tested,
That is characteristic area.For example, pixel threshold value is set as 240, then grey scale pixel value in bank note to be detected is uniform greater than 240
Region is determined as the fisrt feature region of bank note to be detected.
S120, according to the vertical features value, the horizontal properties value and predetermined level equation group, determine described to be checked
The new and old grade of bank note is surveyed, wherein the equation quantity for including in the predetermined level equation group, with the vertical features value and institute
The total quantity for stating horizontal properties value is equal.
Specifically, according to the analysis to high-volume bank note to be detected it is found that the new and old grade point of bank note to be detected with it is vertical
Characteristic value (either horizontal properties value) is linear or non-linear relation, for example, the RMB that face amount is 100 yuan, new and old etc.
Grade value is linear with vertical features value, with horizontal properties value at another linear relationship.Therefore, it is wrapped in predetermined level equation group
The equation quantity contained is equal with the total quantity of vertical features value and horizontal properties value.According to above-mentioned Sobel Operator template switch
At least one available vertical features value and at least one horizontal properties value, are solved further according to preset grade equation group
Multiple preliminary grade points of available bank note to be detected, then mean value is taken to multiple preliminary grade points, it is carried out with level threshold value
Matching, can find out the new and old grade of bank note to be detected.
Wherein, predetermined level equation group be the equation group comprising multiple equations, each equation have it is one's own not
Know parameter and known parameters, that is to say, that the unknown parameter that any one equation calculation goes out does not meet its in equation group simultaneously
His equation.In embodiments of the present invention, unknown parameter be at least one vertical features value or at least one horizontal properties value, and
Bringing the result that unknown parameter is found out into is then bank note to be detected preliminary grade point corresponding with the unknown number, and predetermined level equation
The known parameters of each equation are the specific value for being calculated and being determined previously according to sample bank note in group.
The embodiment of the present invention is based on Sobel Operator template and obtains vertical features value and horizontal properties value, in conjunction with default etc.
Grade equation group determines the new and old grade of bank note to be detected, solves the problems, such as that the new and old recognition effect of bank note is poor in the prior art,
Improve the accuracy of the new and old identification of bank note.
Embodiment two
Fig. 2A is a kind of flow chart for detecting the new aging method of bank note provided by Embodiment 2 of the present invention, and the present embodiment is above-mentioned
On the basis of embodiment further include: determine the pixel characteristic value at least one second feature region in the bank note to be detected；Phase
It answers, according to the vertical features value, the horizontal properties value and predetermined level equation group, determines the bank note to be detected
New and old grade includes: according to the vertical features value, the horizontal properties value, the pixel characteristic value and predetermined level equation
Group determines the new and old grade of the bank note to be detected, and described wherein the equation quantity for including in the predetermined level equation group
The total quantity of vertical features value, the horizontal properties value and the pixel characteristic value is equal.
Correspondingly, the method for the present embodiment includes:
S210, it is based on Sobel Operator template, to the pixel grey scale at least one fisrt feature region in bank note to be detected
Value is converted, and the vertical features value and horizontal properties value at least one fisrt feature region are obtained.
S220, the pixel characteristic value for determining at least one second feature region in the bank note to be detected.
The gray value of pixel each in second feature region is taken mean value as the by the image for acquiring second feature region
The pixel characteristic value of two characteristic areas.Pixel characteristic value equally can be one or more, and number and second feature region are selected
The number taken is related.
Bank note to be detected can also choose at least one second feature region other than choosing fisrt feature region.Example
Such as, the region that grey scale pixel value in bank note to be detected is greater than pixel threshold value can be determined as to the fisrt feature of bank note to be detected
Region and second feature region.
It is preferentially that area is big if having chosen fisrt feature region and second feature region simultaneously in bank note to be detected
Characteristic area as fisrt feature region, the small region of area is as second feature region.Versions in 2005 as shown in Figure 2 B
In the radiation of visible light figure of this 100 yuan of face amount RMB, selecting left side white space 20, (grey scale pixel value is greater than pixel thresholding
Value, and area is big) it is used as fisrt feature region, selecting right side blank region 21, (grey scale pixel value is greater than pixel threshold value, and face
Product is small) it is used as second feature region.Since bank note edge contacts more frequently with finger, the region at bank note edge is set as
Characteristic area is more accurate.
It should be noted that the bank note to be detected in the embodiment of the present invention chooses fisrt feature region and second feature simultaneously
Region is only a kind of example, in other embodiments of the invention, can also only choose the second feature in bank note to be detected
Region identifies the new and old grade of bank note.
S230, according to the vertical features value, the horizontal properties value, the pixel characteristic value and predetermined level equation
Group determines the new and old grade of the bank note to be detected, and described wherein the equation quantity for including in the predetermined level equation group
The total quantity of vertical features value, the horizontal properties value and the pixel characteristic value is equal.
Specifically, if having chosen fisrt feature region and second feature region simultaneously in bank note to be detected, it presets
Grade equation group is then to be worth related equation group, and predetermined level equation with vertical features value, horizontal properties value, pixel characteristic
The equation quantity for including in group, it is equal with the total quantity of vertical features value, horizontal properties value and pixel characteristic value, and predetermined level
Each equation in equation group is related with a characteristic value.
In embodiments of the present invention, the building of the predetermined level equation group include: based on the Sobel Operator template,
The grey scale pixel value at least one fisrt feature region in each sample bank note is converted, each sample bank note is obtained
First sample characteristic value；Determine second sample characteristics at least one second feature region in each sample bank note；
According to the new and old grade point of reality of each sample bank note, the first sample characteristic value and the second sample characteristics structure
Build the predetermined level equation group.
Specifically, the sample banknote image of each new and old grade of acquisition, chooses at least one in each sample bank note respectively
The grey scale pixel value in a fisrt feature region, converts it based on Sobel Operator template, obtains each sample bank note
First sample characteristic value.Wherein, first sample characteristic value includes first sample vertical features value and first sample horizontal properties value.
The grey scale pixel value for choosing at least one second feature region in each sample bank note respectively takes mean value as each sample paper
Second sample characteristics of coin, according to the new and old grade point of reality, first sample characteristic value and the second sample of each sample bank note
Characteristic value constructs predetermined level equation group.
For 100 yuan of face amount RMB of version in 2005 shown in Fig. 2 B, due to according to high-volume bank note to be detected
It is found that the RMB that face amount is 100 yuan, new and old grade point and characteristic value are linear for analysis, therefore construct predetermined level
Equation group are as follows:
Choose the biggish new and old grade of new and old difference be sixty percent newly with ninety percent new sample bank note, practical new and old grade point
It is then respectively 60 and 90.First sample characteristic value by being converted to sixty percent new samples bank note is m1With m2, the second sample characteristics
Value is m3；The first sample characteristic value of ninety percent new samples bank note is m4With m5, the second sample characteristics is m6, bring into respectively above-mentioned
Equation group are as follows:
According to equation group (2) and equation group (3), solution obtains k1, k2, k3, b1, b2With b3Value.
Alternatively, choosing the sample bank note of each new and old grade, it is converted.Obtain multiple first sample characteristic values with
Multiple second sample characteristics reuse least square method algorithm and calculate k1, k2, k3, b1, b2With b3Value.
It should be noted that the building of predetermined level equation group and the reality of each sample bank note are new in the embodiment of the present invention
Old grade point, first sample characteristic value and the second sample characteristics are You Guan only a kind of example, in other implementations of the invention
In example, the building of predetermined level equation group can also be related with practical new and old grade point, first sample characteristic value, alternatively, default
The building of grade equation group can also be related with practical new and old grade point, the second sample characteristics.
The embodiment of the present invention passes through the pixel characteristic value for determining at least one second feature region in bank note to be detected, according to
Vertical features value, horizontal properties value, pixel characteristic value determine the new and old grade of bank note to be detected, so that bank note to be detected is new and old
The identification of grade is more accurate.
Embodiment three
Fig. 3 is a kind of flow chart that detection bank note is new and old that the embodiment of the present invention three provides, and the present embodiment is in above-mentioned implementation
Optimized on the basis of example, provide optimization based on Sobel Operator template, in bank note to be detected at least one first
The grey scale pixel value of characteristic area is converted, and vertical features value and the level for obtaining at least one fisrt feature region are special
The processing method of value indicative, specifically: being multiple ropes by least one fisrt feature region division described in the bank note to be detected
Bell's subregion, and determine the grey scale pixel value of each Sobel subregion；According to each Sobel subregion grey scale pixel value with
And Sobel Operator vertical formwork, determine the vertical features value of each Sobel subregion；By the vertical spy of each Sobel subregion
Vertical features value of the mean value of value indicative as at least one fisrt feature region；Pixel ash according to each Sobel subregion
Angle value and Sobel Operator horizontal shuttering determine the horizontal properties value of each Sobel subregion；By each Sobel subregion
Horizontal properties value of the mean value of horizontal properties value as at least one fisrt feature region.
Correspondingly, the method for the present embodiment includes:
S310, by least one fisrt feature region division described in the bank note to be detected be multiple Sobel sub-districts
Domain, and determine the grey scale pixel value of each Sobel subregion.
Specifically, Sobel Operator according to above and below pixel, left and right adjoint point intensity-weighted it is poor, edge reach extreme value this
One phenomenon detects edge, has smoothing effect to noise.According to Sobel Operator method, the matrix that Sobel template is 3 × 3,
Therefore, be also by the Sobel subregion that the grey scale pixel value in fisrt feature region divides 3 × 3 matrix, and Sobel sub-district
Each point in domain is grey scale pixel value.
S320, grey scale pixel value and Sobel Operator vertical formwork according to each Sobel subregion, determine each rope shellfish
The vertical features value of your subregion.
Wherein, Sobel Operator vertical formwork is as follows:
| -1 | 0 | 1 |
| -2 | 0 | 2 |
| -1 | 0 | 1 |
Each Sobel subregion is multiplied with above-mentioned Sobel vertical formwork respectively, can be obtained and belong to each Sobel sub-district
The vertical features value in domain.
S330, using the mean value of the vertical features value of each Sobel subregion as at least one fisrt feature region
Vertical features value.
Since fisrt feature region is segmented into multiple Sobel subregions, each Sobel subregion hangs down by Sobel
A vertical features value can be all obtained after straight template switch, therefore takes mean value to make the vertical features value of all Sobel subregions
For the vertical features value in fisrt feature region.
S340, grey scale pixel value and Sobel Operator horizontal shuttering according to each Sobel subregion, determine each rope shellfish
The horizontal properties value of your subregion.
Wherein, Sobel Operator horizontal shuttering is as follows:
| 1 | 2 | 1 |
| 0 | 0 | 0 |
| -1 | -2 | -1 |
Each Sobel subregion is multiplied with above-mentioned Sobel horizontal shuttering respectively, can be obtained and belong to each Sobel sub-district
The horizontal properties value in domain.
S350, using the mean value of the horizontal properties value of each Sobel subregion as at least one fisrt feature region
Horizontal properties value.
Specifically, horizontal properties value is identical as the principle of vertical features value, each Sobel subregion passes through Sobel water
A horizontal properties value can be all obtained after flat die plate conversion, therefore takes mean value to make the horizontal properties value of all Sobel subregions
For the horizontal properties value in fisrt feature region.
S360, according to the vertical features value, the horizontal properties value and predetermined level equation group, determine described to be checked
The new and old grade of bank note is surveyed, wherein the equation quantity for including in the predetermined level equation group, with the vertical features value and institute
The total quantity for stating horizontal properties value is equal.
The embodiment of the present invention is by being based on Sobel Operator vertical formwork and Sobel Operator horizontal shuttering, by least one
Fisrt feature region division is multiple Sobel subregions, determines the vertical features value and level at least one fisrt feature region
Characteristic value improves the accuracy of the new and old grade identification of bank note to be detected.
Example IV
Fig. 4 is a kind of flow chart for new aging method of detection bank note that the embodiment of the present invention four provides, and the present embodiment is above-mentioned
Be optimized on the basis of embodiment, provide optimization according to the vertical features value, the horizontal properties value and pre-
If grade equation group, the processing method of the new and old grade of the bank note to be detected is determined, specifically: according to the vertical features
The corresponding preliminary grade of the predetermined level equation group is calculated in value, the horizontal properties value and predetermined level equation group
Value, takes the mean value of each preliminary grade point as the new and old grade point of the bank note to be detected；By the new and old grade point and grade
Threshold value is matched, to obtain the new and old grade of the bank note to be detected.
Correspondingly, the method for the present embodiment includes:
S410, it is based on Sobel Operator template, to the pixel grey scale at least one fisrt feature region in bank note to be detected
Value is converted, and the vertical features value and horizontal properties value at least one fisrt feature region are obtained.
S420, be calculated according to the vertical features value, the horizontal properties value and predetermined level equation group it is described
The corresponding preliminary grade point of predetermined level equation group takes the mean value of each preliminary grade point as the new and old etc. of the bank note to be detected
Grade value.
Specifically, bring vertical features value and horizontal properties value into predetermined level equation group, can be obtained and predetermined level
The corresponding multiple preliminary grade points of equation group.It should be noted that if bank note to be detected is extracted second feature region, then will
Vertical features value, horizontal properties value and pixel characteristic value bring predetermined level equation group into.For example, above-mentioned 100 yuan of version in 2005
The predetermined level equation group of the RMB of face amount is formula (1), therefore, by vertical features value, horizontal properties value and pixel characteristic
Value brings formula (1) into, and y can be obtained1, y2With y3Three preliminary grade points.Three preliminary grade points are averaged again, then may be used
New and old grade point as the bank note to be detected.
S430, the new and old grade point is matched with level threshold value, to obtain the new and old of the bank note to be detected
Grade.
Specifically, the new and old grade of bank note to be detected can according to need be divided into it is multiple, the number of level threshold value with
The new and old number of degrees divided is related.If level threshold value is 5,4 new and old grades are divided；If level threshold value is 11
It is a, then divide 10 new and old grades.For example, threshold value is respectively 100,90,80,70,60,50,40,30,20,10,0, respectively
Corresponding new and old grade be it is ninety percent new, most probably it is new, seventy percent new, sixty percent new, fifty percent new, four Cheng Xin, three Cheng Xin, twenty percent are new, one one-tenth
New and old coin.In another example if being 95 by the new and old grade point for the bank note to be detected being calculated, bank note to be detected
New and old grade be it is ninety percent new；If the new and old grade point by the bank note to be detected being calculated is 12, bank note to be detected
New and old grade be a Cheng Xin.
The embodiment of the present invention is new and old by the way that vertical features value, horizontal properties value and predetermined level equation group to be calculated
Grade point, then new and old grade point is matched with level threshold value, to obtain the new and old grade of bank note to be detected, improve to
Detect the accuracy of bank note.
Embodiment five
Fig. 5 is a kind of structural schematic diagram for new and old device of detection bank note that the embodiment of the present invention five provides, and the present embodiment can
The case where detection new and old suitable for various bank note, this method can be by the devices of the new and old detection of bank note provided in an embodiment of the present invention
It executes, the mode which can be used software and/or hardware is realized, which can be integrated in the new and old detection of any offer bank note
It in the equipment of function, such as can be ATM, paper currency sorter, specifically include: characteristic value acquisition module 51 and level determination module
52。
Characteristic value acquisition module 51, for being based on Sobel Operator template, at least one first spy in bank note to be detected
The grey scale pixel value in sign region is converted, and the vertical features value and horizontal properties at least one fisrt feature region are obtained
Value；
Level determination module 52, for according to the vertical features value, the horizontal properties value and predetermined level equation
Group determines the new and old grade of the bank note to be detected, and described wherein the equation quantity for including in the predetermined level equation group
The total quantity of vertical features value and the horizontal properties value is equal.
On the basis of the above embodiments, further includes: pixel characteristic value obtains module 53.
Pixel characteristic value obtains module 53, for determining the picture at least one second feature region in the bank note to be detected
Plain characteristic value；
The level determination module 52 is specifically used for: according to the vertical features value, the horizontal properties value, the pixel
Characteristic value and predetermined level equation group determine the new and old grade of the bank note to be detected, wherein the predetermined level equation group
In include equation quantity, it is equal with the total quantity of the vertical features value, the horizontal properties value and the pixel characteristic value.
On the basis of the above embodiments, the characteristic value acquisition module 51 is specifically used for: will be in the bank note to be detected
At least one described fisrt feature region division is multiple Sobel subregions, and determines the pixel grey scale of each Sobel subregion
Value；According to the grey scale pixel value and Sobel Operator vertical formwork of each Sobel subregion, each Sobel subregion is determined
Vertical features value；Using the mean value of the vertical features value of each Sobel subregion hanging down as at least one fisrt feature region
Straight characteristic value；According to the grey scale pixel value and Sobel Operator horizontal shuttering of each Sobel subregion, each Sobel is determined
The horizontal properties value in region；Using the mean value of the horizontal properties value of each Sobel subregion as at least one described fisrt feature area
The horizontal properties value in domain.
It on the basis of the above embodiments, further include grade equation group building module 54.
The grade equation group building module 54 is specifically used for: based on the Sobel Operator template, to each sample paper
The grey scale pixel value at least one fisrt feature region is converted in coin, and the first sample for obtaining each sample bank note is special
Value indicative；Determine second sample characteristics at least one second feature region in each sample bank note；According to described each
The new and old grade point of the reality of sample bank note, the first sample characteristic value and second sample characteristics building are described default etc.
Grade equation group.
On the basis of the above embodiments, the level determination module 52 is specifically used for: according to the vertical features value, institute
It states horizontal properties value and the corresponding preliminary grade point of the predetermined level equation group is calculated in predetermined level equation group, take each
New and old grade point of the mean value of preliminary grade point as the bank note to be detected；By the new and old grade point and level threshold value into
Row matching, to obtain the new and old grade of the bank note to be detected.
On the basis of the above embodiments, grey scale pixel value in the bank note to be detected is greater than to the region of pixel threshold value
It is determined as the characteristic area of the bank note to be detected.
The detection new and old device of bank note is for executing the new and old side of detection bank note described in the various embodiments described above described in the present embodiment
The technical effect of method, technical principle and generation is similar, and which is not described herein again.
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of method that detection bank note is new and old characterized by comprising
Based on Sobel Operator template, the grey scale pixel value at least one fisrt feature region in bank note to be detected is turned
It changes, obtains the vertical features value and horizontal properties value at least one fisrt feature region；
According to the vertical features value, the horizontal properties value and predetermined level equation group, the bank note to be detected is determined
New and old grade, wherein the equation quantity for including in the predetermined level equation group, with the vertical features value and the horizontal spy
The total quantity of value indicative is equal；
Wherein, described to be based on Sobel Operator template, to the pixel grey scale at least one fisrt feature region in bank note to be detected
Value is converted, and the vertical features value and horizontal properties value at least one fisrt feature region are obtained, comprising:
It is multiple Sobel subregions by least one fisrt feature region division described in the bank note to be detected, and determines each
The grey scale pixel value of Sobel subregion；
According to the grey scale pixel value and Sobel Operator vertical formwork of each Sobel subregion, each Sobel subregion is determined
Vertical features value；
Using the mean value of the vertical features value of each Sobel subregion as the vertical features at least one fisrt feature region
Value；
According to the grey scale pixel value and Sobel Operator horizontal shuttering of each Sobel subregion, each Sobel subregion is determined
Horizontal properties value；
Using the mean value of the horizontal properties value of each Sobel subregion as the horizontal properties at least one fisrt feature region
Value.
2. the method according to claim 1, wherein further including at least one in the determining bank note to be detected
The pixel characteristic value of two characteristic areas；
According to the vertical features value, the horizontal properties value and predetermined level equation group, the bank note to be detected is determined
New and old grade, comprising:
According to the vertical features value, the horizontal properties value, the pixel characteristic value and predetermined level equation group, institute is determined
The new and old grade of bank note to be detected is stated, wherein the equation quantity for including in the predetermined level equation group, with the vertical features
The total quantity of value, the horizontal properties value and the pixel characteristic value is equal.
3. according to the method described in claim 2, it is characterized in that, the building of the predetermined level equation group includes:
Based on the Sobel Operator template, to the grey scale pixel value at least one fisrt feature region in each sample bank note into
Row conversion, obtains the first sample characteristic value of each sample bank note；
Determine second sample characteristics at least one second feature region in each sample bank note；
According to the new and old grade point of reality of each sample bank note, the first sample characteristic value and second sample characteristics
Value constructs the predetermined level equation group.
4. the method according to claim 1, wherein described according to the vertical features value, the horizontal properties
Value and predetermined level equation group, determine the new and old grade of the bank note to be detected, comprising:
The predetermined level side is calculated according to the vertical features value, the horizontal properties value and predetermined level equation group
The corresponding preliminary grade point of journey group, takes the mean value of each preliminary grade point as the new and old grade point of the bank note to be detected；
The new and old grade point is matched with level threshold value, to obtain the new and old grade of the bank note to be detected.
5. the method according to claim 1, wherein grey scale pixel value in the bank note to be detected is greater than pixel
The region of threshold value is determined as the characteristic area of the bank note to be detected.
6. a kind of device that detection bank note is new and old characterized by comprising
Characteristic value acquisition module, for being based on Sobel Operator template, at least one fisrt feature region in bank note to be detected
Grey scale pixel value converted, obtain the vertical features value and horizontal properties value at least one fisrt feature region；
Level determination module, for determining according to the vertical features value, the horizontal properties value and predetermined level equation group
The new and old grade of the bank note to be detected, wherein the equation quantity for including in the predetermined level equation group, with the vertical spy
The total quantity of value indicative and the horizontal properties value is equal；
Wherein, the characteristic value acquisition module, is specifically used for:
It is multiple Sobel subregions by least one fisrt feature region division described in the bank note to be detected, and determines each
The grey scale pixel value of Sobel subregion；
According to the grey scale pixel value and Sobel Operator vertical formwork of each Sobel subregion, each Sobel subregion is determined
Vertical features value；
Using the mean value of the vertical features value of each Sobel subregion as the vertical features at least one fisrt feature region
Value；
According to the grey scale pixel value and Sobel Operator horizontal shuttering of each Sobel subregion, each Sobel subregion is determined
Horizontal properties value；
Using the mean value of the horizontal properties value of each Sobel subregion as the horizontal properties at least one fisrt feature region
Value.
7. device according to claim 6, which is characterized in that further include:
Pixel characteristic value obtains module, for determining the pixel characteristic at least one second feature region in the bank note to be detected
Value；
The level determination module is specifically used for: according to the vertical features value, the horizontal properties value, the pixel characteristic value
And predetermined level equation group, the new and old grade of the bank note to be detected is determined, wherein including in the predetermined level equation group
Equation quantity, it is equal with the total quantity of the vertical features value, the horizontal properties value and the pixel characteristic value.
8. device according to claim 7, which is characterized in that construct module including grade equation group, be specifically used for:
Based on the Sobel Operator template, to the grey scale pixel value at least one fisrt feature region in each sample bank note into
Row conversion, obtains the first sample characteristic value of each sample bank note；
Determine second sample characteristics at least one second feature region in each sample bank note；
According to the new and old grade point of reality of each sample bank note, the first sample characteristic value and second sample characteristics
Value constructs the predetermined level equation group.
9. device according to claim 6, which is characterized in that the level determination module is specifically used for:
The predetermined level side is calculated according to the vertical features value, the horizontal properties value and predetermined level equation group
The corresponding preliminary grade point of journey group, takes the mean value of each preliminary grade point as the new and old grade point of the bank note to be detected；
The new and old grade point is matched with level threshold value, to obtain the new and old grade of the bank note to be detected.
10. device according to claim 6, which is characterized in that grey scale pixel value in the bank note to be detected is greater than picture
The region of plain threshold value is determined as the characteristic area of the bank note to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710103550.XA CN106910276B (en) | 2017-02-24 | 2017-02-24 | Detect the new and old method and device of bank note |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710103550.XA CN106910276B (en) | 2017-02-24 | 2017-02-24 | Detect the new and old method and device of bank note |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106910276A CN106910276A (en) | 2017-06-30 |
| CN106910276B true CN106910276B (en) | 2019-04-26 |
Family
ID=59208418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710103550.XA Active CN106910276B (en) | 2017-02-24 | 2017-02-24 | Detect the new and old method and device of bank note |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106910276B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109658586B (en) * | 2018-11-20 | 2021-04-23 | 广州广电运通金融电子股份有限公司 | Method and device for calibrating number of sheet-like stackable objects and computer equipment |
| CN115083066B (en) * | 2022-07-20 | 2022-12-06 | 恒银金融科技股份有限公司 | Method and device for detecting whether paper currency is old or new based on digital image |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102682514B (en) * | 2012-05-17 | 2014-07-02 | 广州广电运通金融电子股份有限公司 | Paper identification method and relative device |
| CN105184950A (en) * | 2015-06-03 | 2015-12-23 | 深圳怡化电脑股份有限公司 | Method and device for analyzing banknote to be old or new |
| CN104966349A (en) * | 2015-06-30 | 2015-10-07 | 新达通科技股份有限公司 | ATM platform-based method of detecting new bank note and old bank note |
| CN105957238B (en) * | 2016-05-20 | 2019-02-19 | 聚龙股份有限公司 | A kind of paper currency management method and its system |
| CN106056751B (en) * | 2016-05-20 | 2019-04-12 | 聚龙股份有限公司 | The recognition methods and system of serial number |
-
2017
- 2017-02-24 CN CN201710103550.XA patent/CN106910276B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106910276A (en) | 2017-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114627111B (en) | Textile defect detection and identification device | |
| CN108416766B (en) | Double-side light-entering type light guide plate defect visual detection method | |
| WO2013170663A1 (en) | Paper identifying method and related device | |
| CN106355739B (en) | A kind of method and device that detection bank note is new and old | |
| CN113643294A (en) | Textile defect self-adaptive detection method based on frequency spectrum analysis | |
| CN104504802B (en) | Detection method for paper money splicing signal | |
| CN109523505A (en) | A kind of ceramic brick surface patterned defect detection method based on machine vision | |
| CN106910276B (en) | Detect the new and old method and device of bank note | |
| US20040131242A1 (en) | Monitoring method | |
| CN116977358A (en) | Visual auxiliary detection method for corrugated paper production quality | |
| CN116008289B (en) | Nonwoven product surface defect detection method and system | |
| CN102081045B (en) | Structural damage identification method based on laser television holographic technique | |
| CN116402742A (en) | Visual detection method and system for surface defects of automobile sheet metal part | |
| CN117764990A (en) | method for detecting stamping quality of chassis | |
| CN107862689A (en) | Leather surface substantially damaged automatic identifying method and computer-readable recording medium | |
| CN110059573A (en) | Wild ginseng based on image recognition is classified calibration method | |
| US10388011B2 (en) | Real-time, full web image processing method and system for web manufacturing supervision | |
| CN109035245B (en) | Nondestructive radiographic film sensitivity discrimination method based on gray scale change characteristic analysis | |
| CN106296975A (en) | The recognition methods of a kind of dollar bill face amount and device | |
| EP3459045B1 (en) | Real-time, full web image processing method and system for web manufacturing supervision | |
| Moëll et al. | Comparison of segmentation methods for digital image analysis of confocal microscope images to measure tracheid cell dimensions | |
| Raunio | Quality characterization of tissue and newsprint paper based on image measurements; possibilities of on-line imaging | |
| CN107389677B (en) | Method and device for detecting quality of flannelette fluff | |
| KR101158329B1 (en) | Apparatus and Method for Extracting Fluorescence Pattern for Automatic Paper Money Inspection | |
| JP3322976B2 (en) | Apparatus for determining the degree of contamination of printed matter |
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 |