CN107393126B - A kind of method, apparatus, equipment and the storage medium of bank note version classification - Google Patents
A kind of method, apparatus, equipment and the storage medium of bank note version classification Download PDFInfo
- Publication number
- CN107393126B CN107393126B CN201710565902.3A CN201710565902A CN107393126B CN 107393126 B CN107393126 B CN 107393126B CN 201710565902 A CN201710565902 A CN 201710565902A CN 107393126 B CN107393126 B CN 107393126B
- Authority
- CN
- China
- Prior art keywords
- row
- column
- image
- target image
- bank note
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D11/00—Devices accepting coins; Devices accepting, dispensing, sorting or counting valuable papers
- G07D11/50—Sorting or counting valuable papers
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses method, apparatus, equipment and the storage mediums of a kind of bank note version classification.This method comprises: obtaining the corresponding target image of bank note target signature region to be identified；Whether detect in the target image includes braille feature；If including the bank note to be identified is determined as braille millboard coin；If not including, the bank note to be identified is determined as non-braille millboard coin.Scheme provided by the invention solves the problems, such as the prior art, and by manual identified bank note version, there are high labor cost, low efficiency and identification error rate are high, realization automatically classifies to the version of bank note to be identified by braille feature, improves the efficiency and accuracy rate of detection.
Description
Technical field
The present invention relates to paper money recognition technical field more particularly to a kind of method, apparatus of bank note version classification, equipment and
Storage medium.
Background technique
With the development of economy, the circulation of bank note is increasing.For the ease of arranging a large amount of bank note, need to difference
The bank note of version carries out taxonomic revision.
The identification technology of existing bank note version is mainly based upon manual identified, by dividing the identical bank note of face amount automatically
After the completion of class, then the artificial classification for carrying out version.
However, needing to waste a large amount of human cost by the method manually classified, and imitated by manual identified
Rate is lower, and error rate is higher.
Summary of the invention
The present invention provides method, apparatus, equipment and the storage medium of a kind of bank note version classification, automatically will be wait know with realization
The version of other bank note is classified.
In a first aspect, the embodiment of the invention provides a kind of methods of bank note version classification, this method comprises:
Obtain the corresponding target image of bank note target signature region to be identified；
Whether detect in the target image includes braille feature；
If including the bank note to be identified is determined as braille millboard coin；
If not including, the bank note to be identified is determined as non-braille millboard coin.
Second aspect, the embodiment of the invention also provides a kind of device of bank note version classification, which includes:
First object image collection module, for obtaining the corresponding target image of bank note target signature region to be identified；
Braille feature detection module, for whether detecting in the target image including braille feature；
Braille millboard coin determining module, if for including braille feature in the target image, by the paper to be identified
Coin is determined as braille millboard coin；
Non- braille millboard coin determining module, if for not including braille feature in the target image, by described wait know
Other bank note is determined as non-braille millboard coin.
The third aspect, the embodiment of the invention also provides a kind of electronic equipment, which includes:
One or more processors；
Memory, for storing one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processing
The method that device realizes the bank note version classification that any embodiment of that present invention provides.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program realizes the method for the bank note version classification that any embodiment of that present invention provides when the program is executed by processor.
The present invention is by obtaining the corresponding target image of bank note target signature region to be identified；Detect target image in whether
Including braille feature；If including bank note to be identified is determined as braille millboard coin；It is if not including, bank note to be identified is true
It is set to non-braille millboard coin, solving the prior art, there are high labor cost, low efficiency and knowledges by manual identified bank note version
The high problem of other error rate, realization are automatically classified the version of bank note to be identified by braille feature, improve the effect of detection
Rate and accuracy rate.
Detailed description of the invention
Fig. 1 is the flow chart of the method for one of embodiment of the present invention one bank note version classification；
Fig. 2 is the flow chart of the method for one of embodiment of the present invention two bank note version classification；
Fig. 3 a is the ultraviolet light reflected image of 100 pesos of Cuba's coin of the version in 2000 in the embodiment of the present invention two；
Fig. 3 b is the ultraviolet light reflected image of 100 pesos of Cuba's coin of the version in 2004 in the embodiment of the present invention two；
Fig. 4 a is that 100 pesos of Cuba's coin target signature regions of the version in 2000 in the embodiment of the present invention two are corresponding ultraviolet
Light reflected image；
Fig. 4 b is that 100 pesos of Cuba's coin target signature regions of the version in 2004 in the embodiment of the present invention two are corresponding ultraviolet
Light reflected image；
Fig. 5 a is the corresponding histogram equalization figure of ultraviolet light reflected image shown in Fig. 4 a in the embodiment of the present invention two
Picture；
Fig. 5 b is the corresponding histogram equalization figure of ultraviolet light reflected image shown in Fig. 4 b in the embodiment of the present invention two
Picture；
Fig. 6 a is the corresponding binary image of target image shown in Fig. 5 a in the embodiment of the present invention two；
Fig. 6 b is the corresponding binary image of target image shown in Fig. 5 b in the embodiment of the present invention two；
Fig. 6 c is the target figure that acquisition is further intercepted on the target image shown in Fig. 5 b in the embodiment of the present invention two
Picture；
Fig. 6 d is the corresponding binary image of target image shown in Fig. 6 c in the embodiment of the present invention two；
Fig. 6 e is the target figure that acquisition is further intercepted on the target image shown in Fig. 6 c in the embodiment of the present invention two
Picture；
Fig. 6 f is the corresponding binary image of target image shown in Fig. 6 e in the embodiment of the present invention two；
Fig. 7 is the structural schematic diagram of the device of one of embodiment of the present invention three bank note version classification；
Fig. 8 is the structural schematic diagram of one of the embodiment of the present invention four electronic equipment.
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 of the method for bank note version classification that the embodiment of the present invention one provides, the present embodiment provides
Method be applicable to the case where need to classifying to the version of bank note, can be executed by the device of bank note version classification, the dress
Set can by software and or hardware form, can generally be integrated in the finance devices such as automatic machine or paper money counter.The present embodiment mentions
The method of confession specifically comprises the following steps:
Step 110 obtains the corresponding target image of bank note target signature region to be identified.
Illustratively, the image of bank note to be identified can be obtained by image collecting device first by obtaining target image, then be led to
The mode for crossing setting target signature region coordinate range intercepts the corresponding mesh of target signature region on the image of bank note to be identified
Logo image, and then get target image.For example, preferably the endpoint in the upper left corner is coordinate origin, w table in rectangular coordinate system
Show that abscissa, h indicate ordinate, then in the complete image of bank note to be identified, the position of target signature region may be configured as w=
[a:b] and h=[c:d], and then the rectangular area being made of w=[a:b] and h=[c:d] is obtained as target signature region.Show
Example property, if W is indicated that width, H indicate height, also may specify the top left co-ordinate (e, f) and target of target signature region
The width W and height H of characteristic area, and then the rectangular area being made of (e, f) and W and H is obtained as target signature region.
It should be noted that it includes blind that the currency type of the bank note to be identified in the present embodiment, which can be in any bank note version,
Literary feature, and do not include the currency type of braille feature in another bank note version.Such as bank note to be identified can be RMB, Cuba
The bank note of the currency types such as coin and Thai Baht.
Wherein, target signature region is braille characteristic area.
Preferably, due to the bank note of different currency types corresponding braille characteristic area position difference, such as nineteen ninety-five version RMB
Braille characteristic area be located at the lower right of bank note front positive direction, the braille characteristic area of 2004 Nian Ban Cuba coin is located at bank note
Therefore the upper right side of positive positive direction before the version to bank note to be identified identifies, it is corresponding can first to obtain bank note to be identified
Currency type and bank note to be identified the corresponding bank note of image towards, and then according to currency type, bank note towards with target signature region
The corresponding relationship of coordinate range, determines the target signature region of the bank note to be identified, so that it is special to obtain the bank note target to be identified
Levy the corresponding target image in region.Wherein, bank note is towards including positive positive direction, front opposite direction, reverse side positive direction and anti-
Face opposite direction.
Preferably, since the corresponding braille of the different editions bank note in the bank note of same currency type, including braille feature is special
The position for levying region may be different, and therefore, the target signature region of bank note to be identified is at least one, correspondingly, target signature
The corresponding target image in region is also at least one.Such as different values of money in nineteen ninety-five version RMB and 2015 editions RMB
Braille characteristic area is respectively positioned on the lower right of bank note front positive direction, and in version RMB in 1987 1 yuan, 2 yuan, 5 yuan of value of money or
10 yuan of braille characteristic area is located at the lower left of bank note front positive direction, and value of money is the braille characteristic area position of 50 yuan, 100 yuan
In the lower right of bank note front positive direction, therefore, if bank note to be identified is RMB, target image may include bank note to be identified
The lower left image of positive positive direction and two target images of lower right image of bank note front to be identified positive direction.It is for another example right
In Cuba's coin, the corresponding braille characteristic area of different editions bank note is respectively positioned on the upper right side of bank note front positive direction, therefore, if to
Identification bank note is Cuba's coin, then target image is the upper right side image of bank note front to be identified positive direction.
Illustratively, also the corresponding of bank note to be identified can first be obtained before the version to bank note to be identified identifies
Currency type, value of money and bank note are towards and then according to currency type, value of money and bank note towards pair with target signature region coordinate range
It should be related to, determine the target signature region of the bank note to be identified, so that it is corresponding to obtain the bank note target signature region to be identified
Target image, in such cases, the target signature region of bank note to be identified are one.
It whether include that braille feature if including thens follow the steps 130 in step 120, detection target image, if not including,
Then follow the steps 140.
Bank note to be identified is determined as braille millboard coin by step 130.
Bank note to be identified is determined as non-braille millboard coin by step 140.
After obtaining target image, whether include braille feature, if it is determined that in either objective image if detecting in target image
Include braille feature, then bank note to be identified is determined as braille millboard coin, if it is determined that without braille spy in target image
Sign, then be determined as non-braille millboard coin for bank note to be identified.
Wherein, braille millboard coin be the corresponding currency type of bank note to be identified in include braille feature version bank note, it is non-blind
Literary bank note be the corresponding currency type of bank note to be identified in do not include braille feature version bank note.Such as Cuba's coin, 2000
The version bank note version such as Cuba's coin and 2003 Nian Ban Cuba coin is non-braille millboard coin, 2004 Nian Ban Cuba coin and version in 2014
The bank note version such as Cuba's coin is braille millboard coin.
Method provided in this embodiment as a result, by whether there is braille feature automatically by bank note in detection target image
Version be divided into braille millboard coin and non-two class of braille millboard coin, greatly improve the efficiency and accuracy rate of detection.
Preferably, after bank note to be identified to be divided into two class version bank note of braille millboard coin and non-braille millboard coin, into one
Step can carry out into one each version bank note in braille millboard coin according to the characteristic information of each version bank note in braille millboard coin
Step identification, and according to the characteristic information of each version bank note in non-braille millboard coin, to each version in non-braille millboard coin
Bank note further identified, and then further determines that in braille millboard coin bank note in the version of bank note and non-braille millboard coin
Version.
The present embodiment is by obtaining the corresponding target image of bank note target signature region to be identified；It is in detection target image
No includes braille feature；If including bank note to be identified is determined as braille millboard coin；If not including, by bank note to be identified
Be determined as non-braille millboard coin, solve the prior art by manual identified bank note version there are high labor cost, low efficiency and
Identify the high problem of error rate, realization is automatically classified the version of bank note to be identified by braille feature, improves detection
Efficiency and accuracy rate.
Embodiment two
Fig. 2 be a kind of flow chart of the method for bank note version classification provided by Embodiment 2 of the present invention, the present embodiment be
It is advanced optimized on the basis of above-described embodiment.Method provided in this embodiment specifically comprises the following steps:
Step 210 obtains the corresponding target image of bank note target signature region to be identified.
Preferably, the corresponding target image of bank note target signature region to be identified is obtained, comprising:
Obtain the corresponding ultraviolet light reflected image of bank note target signature region to be identified；
Obtain the corresponding histogram-equalized image of ultraviolet light reflected image.
Wherein, the image of the bank note to be identified of the acquisition under different light can be obtained by image collecting device.Such as may be used
The corresponding infrared light reflection figure of bank note to be identified or ultraviolet light reflectogram are obtained, is not limited this in the present embodiment.
The ultraviolet light reflected image of bank note to be identified is preferably obtained in the present embodiment, and in the ultraviolet light reflected image of acquisition
The corresponding image of upper interception target signature region, and then get the corresponding ultraviolet light reflection of bank note target signature region to be identified
Image.
It illustratively, is 2000 Nian Ban Cuba coin with bank note to be identified in the present embodiment and 2004 Nian Ban Cuba coin are
Example.Fig. 3 a is the ultraviolet light reflected image of 100 pesos of Cuba's coin of version in 2000 provided in this embodiment, and Fig. 3 b is the present embodiment
The ultraviolet light reflected image of 100 pesos of Cuba's coin of the version in 2004 provided.Target signature region is the upper right of bank note to be identified
Square region then intercepts target signature region corresponding ultraviolet light reflected image as shown in fig. 4 a on the basis of Fig. 3 a, is scheming
The corresponding ultraviolet light reflected image of target signature region as shown in Figure 4 b is intercepted on the basis of 3b.
Preferably, after obtaining the corresponding ultraviolet light reflected image of target image, histogram equalization is carried out to the image, with
The variation for improving contrast and gray tone in ultraviolet light reflected image, so that image is more clear.Fig. 5 a is shown in Fig. 4 a
The corresponding histogram-equalized image of ultraviolet light reflected image, Fig. 5 b are the corresponding histogram of ultraviolet light reflected image shown in Fig. 4 b
Figure equalization image.
Preferably, it detects in order to further accurate with the presence or absence of braille feature in target image, then in the target figure of acquisition
The position of braille feature is further accurately located as on the basis of to improve the accuracy of detection braille feature.Wherein, step
220- step 2120 is the accurate process for determining braille feature locations.
Step 220, confirmation are thrown with the presence or absence of the row of any row from top to bottom in the corresponding binary image of target image
Shadow is greater than the 4th preset threshold, if so, step 230 is executed, if it is not, executing step 240.
Row corresponding with row projection is to the corresponding image of last line as target figure in step 230, acquisition target image
Picture, and return to step 220.
After obtaining target image, binaryzation is carried out to target image using P parametric method or other binarization methods, to obtain
The corresponding binary image of target image.Fig. 6 a is the corresponding binary image of target image shown in Fig. 5 a, and Fig. 6 b is Fig. 5 b
Shown in the corresponding binary image of target image.
After obtaining binary image, the row projection of every row first in acquisition binary image, and successively confirm from top to bottom
Row projection with the presence or absence of certain a line is greater than the 4th preset threshold, when the row projection for confirming certain a line is greater than the 4th preset threshold
When, then on the basis of target image, row corresponding with row projection is intercepted in target image to the corresponding image of last line
It as target image, and again returns to and executes step 220, until there is no a certain in the corresponding binary image of target image
Capable row projection is greater than the 4th preset threshold.Hereby it is achieved that further intercept target image, with prevent in target image comprising compared with
The white pixel point of mostly non-braille feature, influences the detection accuracy of subsequent braille feature.
Wherein, the 4th preset threshold can be set according to the actual situation.
Illustratively, the 4th threshold value can be 4/5ths of target image width.Fig. 6 c is the target figure shown in Fig. 5 b
The target image obtained as upper further interception.
After intercepting to target image, then determine the coboundary of braille feature with accurate on the target image after interception
The position of braille feature is positioned, realizes the further interception to target image.Wherein, step 240- step 2120 is further quasi-
Determine the process of the position of position braille feature.
A line in step 240, the corresponding binary image of acquisition target image is as target line.
After obtaining the target image after interception by the above method, binary conversion treatment is carried out to target image, obtains target
The corresponding binary image of image.Due to reduce interception fall pixel influence, if using P parametric method to target image into
Row binary conversion treatment may make binaryzation effect more preferable, be conducive to subsequent more accurately detection braille feature.Meanwhile after interception
Destination image data is reduced, and improves subsequent detection speed.
Fig. 6 d is the corresponding binary image of target image shown in Fig. 6 c.As shown in fig 6d, compared with Fig. 6 b, in Fig. 6 d
More corresponding pixels of blind spot are confirmed as white pixel point.
Step 250, statistics target line start the white pixel point number summation of continuous K row.
Step 260 judges whether number summation is greater than the number summation of preservation, if so, step 270 is executed, if it is not, executing
Step 2100.
Preferably, as shown in fig 6d, it is made of due to braille feature multiple blind spots, and passes through blind spot area after binary conversion treatment
The corresponding pixel in domain is white pixel point.Therefore, each row in binary image can be traversed from top to bottom, and counted
Whether the white pixel point number summation that current traversal row starts continuous K row is greater than the number summation currently saved, if it is determined that
It is greater than, then the current traversal row of explanation may be coboundary, and be verified by step 270 and step 280, if it is determined that less
In then illustrating currently to traverse capable is not coboundary, then traversal to next line, until completing the place to rows all in binary image
Reason.
Wherein, K may be configured as braille feature shared line number in the target image, target image blind spot as shown in fig 6d
Region occupies 13 rows altogether, then K may be configured as 13.
Wherein, number summation initial value may be configured as negative or zero.
Step 270 traverses each column in the first predeterminable area from left to right, counts current traversal column and starts continuous K column
White pixel point number summation records corresponding first column position of maximum number summation in the first predeterminable area, and from left to right
It traverses each column in the second predeterminable area, counts the white pixel point number summation that current traversal column start continuous K column, record the
Corresponding second column position of maximum number summation in two predeterminable areas.
Specifically, the first predeterminable area and the second predeterminable area can be set according to actually detected braille feature,
In the first predeterminable area and the second predeterminable area and two different blind spots in braille feature it is corresponding.
With first left blind spot in first predeterminable area corresponding diagram 6d the second row blind spot, the second predeterminable area corresponding diagram 6d
In second row blind spot for third left blind spot.The height of each blind spot known in advance is n1, i.e., line number shared by each blind spot
For n1, then can by binary image, by currently traversing under row, with current traversal row be separated by row that line number is n1, with it is current
Traversal row is separated by the rectangular area constituted at the half of the row that line number is 2n1, left side first row and binary image width and makees
It for the first predeterminable area, and traverses each column from left to right in the first predeterminable area, counts current traversal column and start continuous K column
White pixel point number summation obtains and respectively arranges corresponding white pixel point number summation in the first predeterminable area, and records first
Corresponding first column position of maximum number summation in predeterminable area, using the first column position as the corresponding column of first left blind spot
Position.And can by binary image, by currently traversing under row, with current traversal row be separated by row that line number is n1, with it is current
Traversal row is separated by line number for the row of 2n1, the right side of the first column position and the column and first for being separated by the first columns with the first column position
The rectangular area that the right side of column position and the column for being separated by the second columns with the first column position are constituted is as the second predeterminable area, and the
Each column are traversed in two predeterminable areas from left to right, the white pixel point number summation that current traversal column start continuous K column is counted, obtains
It takes and respectively arranges corresponding white pixel point number summation in the second predeterminable area, and record maximum number summation in the second predeterminable area
Corresponding second column position, using the column position as the corresponding column position of third left blind spot.
Whether within a preset range step 280 determines the difference of the first column position and the second column position, if so, executing step
290, if it is not, executing step 2100.
The number summation of preservation is updated to currently traverse the corresponding number summation of row by step 290, and by the line position of preservation
It sets and is updated to when row traversal row.
After determining the first column position and the second column position, the difference of the first column position and the second column position is calculated, if it is determined that
The number summation of preservation within a preset range, then is updated to currently traverse row correspondence by the difference of the first column position and the second column position
Number summation, and the line position of preservation set and is updated to when row traversal row.
Wherein, the line position of preservation, which sets initial position, may be configured as sky.
Wherein, preset range can be set according to the actual situation.Such as when first predeterminable area corresponding diagram the second row of 6d is blind
First left blind spot in point, in second predeterminable area corresponding diagram 6d the second row blind spot when third left blind spot, preset range
It can be 17-22.
It can be excluded by step 270- step 290 as in Fig. 6 b, there are more white pictures for certain a line on braille feature
Vegetarian refreshments causes the row of the row to project corresponding white pixel point number summation and is greater than the white saved such as most lastrow in Fig. 6 b
Pixel summation, but the case where row is determined as coboundary, can be excluded by step 270- step 290, meanwhile, if two-value
There are more noises in change image, and the interference of noise can be equally excluded by the above method, greatly improves the standard of determining coboundary
True property.
Step 2100 judges whether to complete the processing to all rows in binary image, if so, step 2110 is executed,
If it is not, executing step 240.
After completing the processing to all rows in binary image, then the line position saved is set as coboundary.
Step 2110 obtains in target image on the line position of preservation is set, and sets that be separated by first default with the line position of preservation
The row of line number is set with the line position of preservation and is separated by the corresponding image conduct of the row of the second default line number under the line position of preservation is set
Target image.
After determining coboundary, braille feature corresponding target image is intercepted on the basis of target image again.Such as interception
The lastrow that the line position saved in target image is set to the line position saved sets the row for differing that line number is K with the line position of preservation under setting
Corresponding image is as target image.
Fig. 6 e is the target image that acquisition is further intercepted on the target image shown in Fig. 6 c.As shown in fig 6e, with figure
6c is compared, and Fig. 6 e intercepts out complete braille feature, while greatly reducing image data, greatly reduces the dry of irrelevant point
It disturbs, to improve the accuracy of detection.
After target image after obtaining interception, whether detection target image includes braille feature, step 2120- step 2240
For the process for detecting braille feature.
Step 2120, in the third predeterminable area of the corresponding binary image of target image, slided with M*N window
It is dynamic, and count white pixel point number in each window.
After obtaining target image, binaryzation is carried out to target image, obtains the corresponding binary image of target image.Fig. 6 f
For the corresponding binary image of target image shown in Fig. 6 e.
It in the third predeterminable area of binary image, is slided with M*N window, and counts white pixel in each window
Point number.
Specifically, third predeterminable area can be set according to actually detected braille feature.Wherein, third predeterminable area
It is corresponding with the first blind spot.
By taking the first blind spot is first left blind spot in binary image the first row blind spot as shown in Figure 6 f as an example.Then
Three predeterminable areas can be the 0th row to the 10th row from top to bottom, and the 0th arrange from left to right to binary image width one third
The rectangular area of composition.
Wherein, M and N can be equal, can not also wait.If M and N can be all 6.
Step 2130, corresponding first left end point of window for saving maximum white pixel point number in third predeterminable area are sat
Cursor position.
The corresponding first left end point coordinate position of window of white pixel point number maximum in third predeterminable area is determined
The Position Approximate where the first blind spot, and record the first left end point coordinate position.
Step 2140 determines the 4th predeterminable area according to the first left end point coordinate position.
The 4th predeterminable area is determined according to the first left end point coordinate position, i.e., determines according to the first left end point coordinate position
Region where one blind spot.If the first left end point coordinate position is (X, Y), then the 4th predeterminable area can be abscissa X-1, horizontal
The rectangular area that coordinate X+9, ordinate Y-2 and ordinate Y+9 are constituted.
Step 2150 projects according to the row projection of the 4th each row of predeterminable area and the respectively column that arrange and determines in binary image the
The row bound of one blind spot and column boundary, and count white pixel point in the region being made of the row bound of the first blind spot and column boundary
First number.
The column projection for calculating the row projection of each row in the 4th predeterminable area and respectively arranging, and the basis in the 4th predeterminable area
Row projection and column project the row bound for determining the first blind spot and column boundary, and wherein row bound includes coboundary and lower boundary, arrange side
Boundary includes left margin and right margin.
Such as coboundary, it can detect whether that the row projection there are continuous two row is greater than default row projection value from top to bottom,
If so, a line for being located at upside in continuous two row is determined as coboundary；For lower boundary, can detect whether to deposit from bottom to up
It is greater than default row projection value in the row projection of continuous two row, if so, a line by continuous two line position in downside is determined as below
Boundary；For left margin, it can detect whether that the column projection there are continuous two column is greater than default column projection value from left to right, if so,
The column for being located at left side in continuous two column are determined as left margin；For right margin, can detect whether to exist from right to left continuous
The column projection of two column is greater than default column projection value, if so, the column for being located at right side in continuous two column are determined as right margin.
After determining row bound and column boundary, statistics is by row bound and arranges the white pixel point in the rectangular area that boundary is constituted
First number.
Step 2160 determines the 5th predeterminable area according to the column boundary of the first blind spot.
Wherein, the 5th predeterminable area is corresponding with the second blind spot.It is binary image as shown in Figure 6 f the with the second blind spot
In a line blind spot for second left blind spot.As the 5th predeterminable area can be for by the 0th row to the 10th row from top to bottom, Yi Ji
The rectangular area constituted at the right margin of one blind spot to picture traverse half.
Step 2170, in the 5th predeterminable area, slided with M*N window, and count white pixel point in each window
Number.
Step 2180, corresponding second left end point of window for saving maximum white pixel point number in the 5th predeterminable area are sat
Cursor position.
Step 2190 determines the 6th predeterminable area according to the second left end point coordinate position.
Step 2200 projects according to the row of the 6th each row of predeterminable area and respectively arranges the corresponding determining binary image of column projection
In the second blind spot row bound and column boundary, and count by the row bound of the second blind spot and the region words spoken by an actor from offstage colour that constitutes of column boundary
Second number of vegetarian refreshments.
Wherein, step 2170- step 2200 determines the row bound of the second blind spot and arranges boundary and statistics by the second blind spot
Row bound and the region that constitutes of column boundary in white pixel point second number and the determination of step 2120- step 2150 it is first blind
White pixel point in the region that the row bound of point and column boundary and statistics are made of the row bound of the first blind spot and column boundary
First several process is identical, repeats no more in the present embodiment.
Whether the difference of the row bound of step 2210, the row bound for determining the first blind spot and the second blind spot is less than the first default threshold
Value and or the first blind spot the column boundary for arranging boundary and the second blind spot difference whether less than the second preset threshold, if so, executing step
Rapid 2220, if it is not, executing step 2240.
Illustratively, if the first blind spot is that first left is blind in the first row blind spot in binary image shown in Fig. 6 f
Point, the second blind spot are second left blind spot in the first row blind spot, then need to only determine the left margin of the first blind spot and the second blind spot
Difference and or right margin difference whether less than the first preset threshold, if the first blind spot is the in binary image shown in Fig. 6 f
Second left blind spot in a line blind spot, the second blind spot are second left blind spot in the second row blind spot, then need to only determine first
The difference of the coboundary of blind spot and the second blind spot and or lower boundary difference whether less than the second preset threshold, if the first blind spot is figure
First left blind spot in the first row blind spot in binary image shown in 6f, the second blind spot are the left side second in the second row blind spot
A blind spot, then need the difference for determining the coboundary of the first blind spot and the second blind spot and or the difference of lower boundary whether preset less than first
The difference of the coboundary of threshold value and the first blind spot and the second blind spot and or lower boundary difference whether less than the second preset threshold.
Wherein, the second preset threshold can be identical as the first preset threshold, can also be different.
Step 2220 determines whether first number and second several difference are less than third predetermined threshold value, if so, executing step
2230, if it is not, executing step 2240.
If it is determined that the difference of the row bound of the row bound of the first blind spot and the second blind spot less than the first preset threshold and or first
The difference on the column boundary on the column boundary and the second blind spot of blind spot is less than the second preset threshold, it is determined that first number and second it is several it
Whether difference is less than third predetermined threshold value, if so, determining in target image includes braille feature, bank note to be identified is determined as blind
Literary millboard coin, if not, it is determined that do not include braille feature in target image, bank note to be identified is determined as non-braille millboard coin.
Step 2230 determines in target image to include braille feature, and bank note to be identified is determined as braille millboard coin.
Step 2240 determines do not include braille feature in target image, and bank note to be identified is determined as non-braille millboard coin.
Method provided in this embodiment can by confirm in the corresponding binary image of target image, from top to bottom whether
There are the row projections of any row to be greater than the 4th preset threshold, and realization further intercepts target image, reduces irrelevant point pair
The influence of subsequent braille feature detection, while the data volume for reducing subsequent processing improves detection speed；And it can realize accurately really
The coboundary for determining braille feature further accurately intercepts target image by coboundary, equally reduction irrelevant
Whether the influence that point detects subsequent braille feature, improves the accuracy of detection, by including that braille is special on detection target image
Sign, if including, is determined as braille millboard coin for bank note to be identified, if not including, bank note to be identified is determined as non-braille
Millboard coin, realization automatically and accurately classify to bank note to be identified by braille feature.
Embodiment three
Fig. 7 is a kind of structural schematic diagram of the device for bank note version classification that the embodiment of the present invention three provides.The device is suitable
The case where for that need to classify to bank note version, which can be made of software and/or hardware, can generally be integrated in automatic selling
In the finance devices such as ticket machine or paper money counter.Device provided in this embodiment includes: first object image collection module 310, braille
Feature detection module 320, braille millboard coin determining module 330 and non-braille millboard coin determining module 340, wherein
First object image collection module 310, for obtaining the corresponding target image of bank note target signature region to be identified；
Braille feature detection module 320, for whether detecting in the target image including braille feature；
Braille millboard coin determining module 330, if for including braille feature in the target image, it will be described to be identified
Bank note is determined as braille millboard coin；
Non- braille millboard coin determining module 340, if for not including braille feature in the target image, will it is described to
Identification bank note is determined as non-braille millboard coin.
Device provided in this embodiment obtains the corresponding mesh of bank note target signature region to be identified by first object image
Logo image；Whether it includes braille feature that braille feature detection module detects in target image；If braille millboard coin determining module mesh
Include braille feature in logo image, then bank note to be identified is determined as braille millboard coin；If non-braille millboard coin determining module mesh
Do not include braille feature in logo image, then bank note to be identified is determined as non-braille millboard coin, solves the prior art and pass through people
Work identification bank note version has that high labor cost, low efficiency and identification error rate are high, realizes automatic by braille feature
The version of bank note to be identified is classified, the efficiency and accuracy rate of detection are improved.
In above scheme, optionally, described device further include:
Number summation statistical module, for obtaining a line in the corresponding binary image of the target image as target
Row, and count the white pixel point number summation that the target line starts continuous K row；
Column position logging modle traverses the if being greater than the number summation saved for the number summation from left to right
Each column in one predeterminable area count the white pixel point number summation that current traversal column start continuous K column, record described first
Corresponding first column position of maximum number summation in predeterminable area, and each column in the second predeterminable area are traversed from left to right, it unites
The current traversal column of meter start the white pixel point number summation of continuous K column, and it is total to record maximum number in second predeterminable area
With corresponding second column position；
The difference determining module of position, for determining the difference of first column position and second column position whether default
In range；
Update module is saved, if within a preset range for the difference of first column position and second column position,
The number summation of preservation is updated to currently to traverse the corresponding number summation of row, and the line position of preservation is set and is updated to when row traversal
Row, and a line executed obtain in the corresponding binary image of the target image is returned as target line；
Return module is executed, if within a preset range for the difference of first column position and second column position,
The a line executed obtain in the corresponding binary image of the target image is returned as target line, until completing to the two-value
Change the processing of all rows in image；
Second target image obtains module, and described for obtaining on the line position in the target image in preservation sets
The line position of preservation, which is set, to be separated by the row of the first default line number and is separated by the to being set under the line position of preservation is set with the line position of the preservation
The corresponding image of row of two default line numbers is as target image.
In above scheme, optionally, the braille feature detection module, comprising:
First several statistic unit, in the third predeterminable area of the corresponding binary image of the target image,
It is slided with M*N window, and counts white pixel point number in each window；
First coordinate position storage unit, for saving the window of maximum white pixel point number in the third predeterminable area
The corresponding first left end point coordinate position of mouth；
4th predeterminable area determination unit, for determining the 4th predeterminable area according to the first left end point coordinate position；
Second several statistic unit, it is true for being projected according to the row projection of each row of the 4th predeterminable area and the column of each column
The row bound of first blind spot and column boundary in the fixed binary image, and count the row bound and Lie Bian by first blind spot
First number of white pixel point in the region that boundary is constituted；
5th predeterminable area determination unit, for determining the 5th predeterminable area according to the column boundary of first blind spot；
Third number statistic unit for being slided in the 5th predeterminable area with M*N window, and counts each
White pixel point number in window；
Second coordinate position storage unit, for saving the window of maximum white pixel point number in the 5th predeterminable area
The corresponding second left end point coordinate position of mouth；
6th predeterminable area determination unit, for determining the 6th predeterminable area according to the second left end point coordinate position；
4th several statistic unit, for being thrown according to the row projection and the corresponding column of each column of each row of the 6th predeterminable area
Shadow determines in the binary image row bound of the second blind spot and column boundary, and count by second blind spot row bound and
Arrange second number of white pixel point in the region that boundary is constituted；
The difference determination unit on boundary, for determine first blind spot row bound and second blind spot row bound it
Difference whether less than the first preset threshold and or first blind spot column boundary and second blind spot column boundary difference whether
Less than the second preset threshold；
The difference determination unit of number, for if it is determined that the row bound of first blind spot and the row bound of second blind spot
Difference less than the first preset threshold and or first blind spot column boundary and second blind spot column boundary difference less than the
Two preset thresholds, it is determined that whether first number and second several difference are less than third predetermined threshold value；
Braille characteristics determining unit is used for if it is determined that first number and second several difference are default less than third
Threshold value, it is determined that include braille feature in the target image；
Non- braille characteristics determining unit is used for if it is determined that first number and second several difference are pre- less than third
If threshold value, it is determined that do not include braille feature in the target image.
In above scheme, optionally, described device further include:
Row projection confirmation module, for confirm in the corresponding binary image of the target image, from top to bottom whether
There are the row projections of any row to be greater than the 4th preset threshold；
Third target image obtains module, if for confirming in the corresponding binary image of the target image, from upper
It is greater than the 4th preset threshold to the lower row projection there are any row, then obtains corresponding with row projection in the target image
It goes to the corresponding image of last line as target image, and returns to execution confirmation in the corresponding binary picture of the target image
As in, it is greater than the 4th preset threshold with the presence or absence of the row projection of any row from top to bottom, until confirmation is in the target image pair
In the binary image answered, there is no the row projections of any row to be greater than the 4th preset threshold from top to bottom.
In above scheme, optionally, the target image is specially histogram-equalized image.
In above scheme, optionally, the first object obtains image module, comprising:
Image acquisition unit, for obtaining the corresponding ultraviolet light reflected image of bank note target signature region to be identified；
Histogram-equalized image acquiring unit, for obtaining the corresponding histogram equalization of the ultraviolet light reflected image
Image.
Example IV
Fig. 8 is the structural schematic diagram for a kind of electronic equipment that the embodiment of the present invention four provides, as shown in figure 8, the electronics is set
Standby includes processor 410, memory 420, input unit 430 and output device 440；The quantity of processor 410 in electronic equipment
It can be one or more, in Fig. 8 by taking a processor 410 as an example；It is processor 410, memory 420 in electronic equipment, defeated
Entering device 430 can be connected with output device 440 by bus or other modes, in Fig. 8 for being connected by bus.
Memory 420 is used as a kind of computer readable storage medium, can be used for storing software program, journey can be performed in computer
Sequence and module, as the corresponding program instruction/module of the method for the bank note version classification in any embodiment of that present invention (for example,
First object image collection module 310, braille feature detection module 320, braille millboard coin in the device of bank note version classification
Determining module 330 and non-braille millboard coin determining module 340).Processor 410 is stored in soft in memory 420 by operation
Part program, instruction and module realize above-mentioned use thereby executing the various function application and data processing of electronic equipment
In the operation of electronic equipment.
Memory 420 can mainly include storing program area and storage data area, wherein storing program area can store operation system
Application program needed for system, at least one function；Storage data area, which can be stored, uses created data according to electronic equipment
Deng.In addition, memory 420 may include high-speed random access memory, it can also include nonvolatile memory, for example, at least
One disk memory, flush memory device or other non-volatile solid state memory parts.In some instances, memory 420 can
It further comprise the memory remotely located relative to processor 410, these remote memories can pass through network connection to electricity
Sub- equipment.The example of above-mentioned network includes but is not limited to internet, intranet, local area network, mobile radio communication and combinations thereof.
Input unit 430 can be used for receiving the number or character information of input, and generates and set with the user of electronic equipment
It sets and the related key signals of function control inputs.Output device 440 may include that display screen etc. shows equipment.
Embodiment five
The embodiment of the present invention five also provides a kind of storage medium comprising computer executable instructions, is stored thereon with calculating
Machine program realizes the method for the bank note version classification that any embodiment of that present invention provides when the program is executed by processor.
By the description above with respect to embodiment, it is apparent to those skilled in the art that, the present invention
It can be realized by software and required common hardware, naturally it is also possible to which by hardware realization, but in many cases, the former is more
Good embodiment.Based on this understanding, technical solution of the present invention substantially in other words contributes to the prior art
Part can be embodied in the form of software products, which can store in computer readable storage medium
In, floppy disk, read-only memory (Read-Only Memory, ROM), random access memory (Random such as computer
Access Memory, RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are with so that a computer is set
Standby (can be personal computer, server or the network equipment etc.) executes method described in any embodiment of that present invention.
It is worth noting that, in the embodiment of the device of above-mentioned bank note version classification, included each unit and module
It is only divided according to the functional logic, but is not limited to the above division, as long as corresponding functions can be realized；
In addition, the specific name of each functional unit is also only for convenience of distinguishing each other, the protection scope being not intended to restrict the invention.
Method provided by any embodiment of the invention can be performed in above-mentioned apparatus, has and executes the corresponding function of the above method
Module and beneficial effect.The not technical detail of detailed description in the present embodiment, reference can be made to any embodiment of that present invention is provided
Method.
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 (9)
1. a kind of method of bank note version classification characterized by comprising
Obtain the corresponding target image of bank note target signature region to be identified；
Whether detect in the target image includes braille feature；
If including the bank note to be identified is determined as braille millboard coin；
If not including, the bank note to be identified is determined as non-braille millboard coin；
It is described obtain the corresponding target image of bank note target signature region to be identified after, further includes:
A line in the corresponding binary image of the target image is obtained as target line, and counts the target line and starts to connect
The white pixel point number summation of continuous K row；
If the number summation is greater than the number summation saved, each column in the first predeterminable area are traversed from left to right, are counted
Current traversal column start the white pixel point number summation of continuous K column, record maximum number summation in first predeterminable area
Corresponding first column position, and each column in the second predeterminable area are traversed from left to right, it counts current traversal column and starts continuous K column
White pixel point number summation, record corresponding second column position of maximum number summation in second predeterminable area；
Whether within a preset range to determine the difference of first column position and second column position；
If so, the number summation of preservation is updated to currently to traverse the corresponding number summation of row, and the line position of preservation is set more
New is to work as row traversal row, and return to a line in the corresponding binary image of the execution acquisition target image as target line；
If it is not, then returning to a line executed obtain in the corresponding binary image of the target image as target line, until complete
The processing of all rows in the pairs of binary image；
It obtains on the line position in the target image in preservation sets, is set with the line position of the preservation and be separated by the first default line number
Row is set with the line position of the preservation and is separated by the corresponding image of the row of the second default line number as mesh under the line position of preservation is set
Logo image.
2. the method according to claim 1, wherein whether including that braille is special in the detection target image
Sign, comprising:
It in the third predeterminable area of the corresponding binary image of the target image, is slided with M*N window, and counts each
White pixel point number in window；
Save the corresponding first left end point coordinate position of window of maximum white pixel point number in the third predeterminable area；
The 4th predeterminable area is determined according to the first left end point coordinate position；
It is blind according in the determining binary image of column projection of the row projection and each column of each row of the 4th predeterminable area first
The row bound of point and column boundary, and count white pixel point in the region being made of the row bound of first blind spot and column boundary
First number；
The 5th predeterminable area is determined according to the column boundary of first blind spot；
It in the 5th predeterminable area, is slided with M*N window, and counts white pixel point number in each window；
Save the corresponding second left end point coordinate position of window of maximum white pixel point number in the 5th predeterminable area；
The 6th predeterminable area is determined according to the second left end point coordinate position；
According in the row projection of each row of the 6th predeterminable area and the determining binary image of each column corresponding column projection the
The row bound of two blind spots and column boundary, and count the region words spoken by an actor from offstage colour being made of the row bound of second blind spot and column boundary
Second number of vegetarian refreshments；
Determine the row bound of first blind spot and the row bound of second blind spot difference whether less than the first preset threshold and
Or whether the difference on the column boundary on the column boundary and second blind spot of first blind spot is less than the second preset threshold；
If so, determining whether first number and second several difference are less than third predetermined threshold value；
If so, determining includes braille feature in the target image；
If not, it is determined that do not include braille feature in the target image.
3. the method according to claim 1, wherein described obtain the corresponding binary image of the target image
In a line as target line, and before counting the white pixel point number summation that the target line starts continuous K row, also wrap
It includes:
Confirmation is greater than the with the presence or absence of the row projection of any row from top to bottom in the corresponding binary image of the target image
Four preset thresholds；
If confirmation exists, obtains row corresponding with row projection to the corresponding image of last line in the target image and make
For target image, and execution confirmation is returned in the corresponding binary image of the target image, from top to bottom with the presence or absence of appointing
A line row projection be greater than the 4th preset threshold, until confirmation in the corresponding binary image of the target image, from up to
There is no the row projections of any row to be greater than the 4th preset threshold down.
4. the method according to claim 1, wherein the target image is specially histogram-equalized image.
5. according to the method described in claim 4, it is characterized in that, the acquisition bank note target signature region to be identified is corresponding
Target image, comprising:
Obtain the corresponding ultraviolet light reflected image of bank note target signature region to be identified；
Obtain the corresponding histogram-equalized image of the ultraviolet light reflected image.
6. the method according to claim 1, wherein the braille millboard coin be 2004 Nian Ban Cuba coin, it is described
Non- braille millboard coin is 2000 Nian Ban Cuba coin.
7. a kind of device of bank note version classification characterized by comprising
First object image collection module, for obtaining the corresponding target image of bank note target signature region to be identified；
Braille feature detection module, for whether detecting in the target image including braille feature；
Braille millboard coin determining module, if for including braille feature in the target image, the bank note to be identified is true
It is set to braille millboard coin；
Non- braille millboard coin determining module, if for not including braille feature in the target image, by the paper to be identified
Coin is determined as non-braille millboard coin；
Number summation statistical module, for obtaining a line in the corresponding binary image of the target image as target line,
And count the white pixel point number summation that the target line starts continuous K row；
It is pre- to traverse first if being greater than the number summation saved for the number summation from left to right for column position logging modle
If each column in region, the white pixel point number summation that current traversal column start continuous K column is counted, it is default to record described first
Corresponding first column position of maximum number summation in region, and each column in the second predeterminable area are traversed from left to right, statistics is worked as
Preceding traversal column start the white pixel point number summation of continuous K column, record maximum number summation pair in second predeterminable area
The second column position answered；
The difference determining module of position, for determining the difference of first column position and second column position whether in preset range
It is interior；
Update module is saved, if within a preset range for the difference of first column position and second column position, will protect
The number summation deposited is updated to currently traverse the corresponding number summation of row, and the line position of preservation is set and is updated to when row traversal row,
And a line executed obtain in the corresponding binary image of the target image is returned as target line；
Return module is executed, if within a preset range for the difference of first column position and second column position, returning
A line in the corresponding binary image of the acquisition target image is executed as target line, until completing to the binary picture
The processing of all rows as in；
Second target image obtains module, for obtaining on the line position in the target image in preservation sets, with the preservation
Line position set and be separated by the row of the first default line number to be separated by second pre- to being set under the line position of preservation is set with the line position of the preservation
If the corresponding image of the row of line number is as target image.
8. a kind of electronic equipment, which is characterized in that the equipment includes:
One or more processors；
Memory, for storing one or more programs；
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method of the bank note version classification as described in claim 1-6 is any.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The method of the bank note version classification as described in claim 1-6 is any is realized when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710565902.3A CN107393126B (en) | 2017-07-12 | 2017-07-12 | A kind of method, apparatus, equipment and the storage medium of bank note version classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710565902.3A CN107393126B (en) | 2017-07-12 | 2017-07-12 | A kind of method, apparatus, equipment and the storage medium of bank note version classification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107393126A CN107393126A (en) | 2017-11-24 |
| CN107393126B true CN107393126B (en) | 2019-10-11 |
Family
ID=60339373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710565902.3A Active CN107393126B (en) | 2017-07-12 | 2017-07-12 | A kind of method, apparatus, equipment and the storage medium of bank note version classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107393126B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117115486B (en) * | 2023-10-19 | 2024-03-19 | 恒银金融科技股份有限公司 | Banknote version identification method and device and electronic equipment |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1639741A (en) * | 2002-08-30 | 2005-07-13 | 富士通先端科技株式会社 | Paper sheets characteristic detection device and paper sheets characteristic detection method |
| CN200997199Y (en) * | 2007-01-24 | 2007-12-26 | 蒋清晓 | Automatic intelligent reader for blind |
| CN201359741Y (en) * | 2008-12-25 | 2009-12-09 | 慕和 | Automatic release machine with identity and bill information acquisition identifier |
| CN101801672A (en) * | 2007-08-14 | 2010-08-11 | 株式会社理光 | Image processing apparatus, image forming apparatus, and image processing method |
| CN101814128A (en) * | 2009-02-20 | 2010-08-25 | 窦鹃鹃 | Portable dollar note denomination identifier for blind |
| CN201707726U (en) * | 2010-05-25 | 2011-01-12 | 李泓 | Paper money value identifying and reading device |
| CN102129800A (en) * | 2010-01-19 | 2011-07-20 | 湛江恒立捷尔达科技有限公司 | System for identifying and automatically phonically prompting blind symbol information |
| CN104783429A (en) * | 2015-04-28 | 2015-07-22 | 百度在线网络技术（北京）有限公司 | Smart wallet and method and device for determining balance |
| CN106447904A (en) * | 2016-09-09 | 2017-02-22 | 深圳怡化电脑股份有限公司 | Paper money identifying method and apparatus |
| CN106446885A (en) * | 2016-09-21 | 2017-02-22 | 北京联合大学 | Paper-based Braille recognition method and system |
| CN106875542A (en) * | 2017-01-09 | 2017-06-20 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note version and its device |
-
2017
- 2017-07-12 CN CN201710565902.3A patent/CN107393126B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1639741A (en) * | 2002-08-30 | 2005-07-13 | 富士通先端科技株式会社 | Paper sheets characteristic detection device and paper sheets characteristic detection method |
| CN200997199Y (en) * | 2007-01-24 | 2007-12-26 | 蒋清晓 | Automatic intelligent reader for blind |
| CN101801672A (en) * | 2007-08-14 | 2010-08-11 | 株式会社理光 | Image processing apparatus, image forming apparatus, and image processing method |
| CN201359741Y (en) * | 2008-12-25 | 2009-12-09 | 慕和 | Automatic release machine with identity and bill information acquisition identifier |
| CN101814128A (en) * | 2009-02-20 | 2010-08-25 | 窦鹃鹃 | Portable dollar note denomination identifier for blind |
| CN102129800A (en) * | 2010-01-19 | 2011-07-20 | 湛江恒立捷尔达科技有限公司 | System for identifying and automatically phonically prompting blind symbol information |
| CN201707726U (en) * | 2010-05-25 | 2011-01-12 | 李泓 | Paper money value identifying and reading device |
| CN104783429A (en) * | 2015-04-28 | 2015-07-22 | 百度在线网络技术（北京）有限公司 | Smart wallet and method and device for determining balance |
| CN106447904A (en) * | 2016-09-09 | 2017-02-22 | 深圳怡化电脑股份有限公司 | Paper money identifying method and apparatus |
| CN106446885A (en) * | 2016-09-21 | 2017-02-22 | 北京联合大学 | Paper-based Braille recognition method and system |
| CN106875542A (en) * | 2017-01-09 | 2017-06-20 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note version and its device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107393126A (en) | 2017-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6878575B2 (en) | Banknote management methods, systems, programs and recording media | |
| TWI619080B (en) | Method for calculating fingerprint overlapping region and electronic device | |
| TWI655586B (en) | Method and device for detecting specific identification image in predetermined area | |
| US9827599B2 (en) | Banknote recognition and classification method and system | |
| CN107437294B (en) | Character segmentation method, device, equipment and storage medium | |
| CN110781885A (en) | Text detection method, device, medium and electronic equipment based on image processing | |
| CN107103683B (en) | Paper money identification method and device, electronic equipment and storage medium | |
| CN106952393B (en) | Paper money identification method and device, electronic equipment and storage medium | |
| CN106372564A (en) | Gesture identification method and apparatus | |
| CN104992496A (en) | Paper money face identification method and apparatus | |
| CN107610316B (en) | Method and device for detecting defect of paper money and terminal equipment | |
| CN107610322A (en) | Recognition methods, device, electronic equipment and the storage medium of bank note version | |
| CN105225523A (en) | A kind of parking space state detection method and device | |
| CN109190639A (en) | A kind of vehicle color identification method, apparatus and system | |
| CN106780966A (en) | A kind of paper money discrimination method and device | |
| CN108734846B (en) | Method and device for denoising paper money image, terminal and storage medium | |
| CN108460775A (en) | A kind of forge or true or paper money recognition methods and device | |
| CN113989626B (en) | Multi-class garbage scene distinguishing method based on target detection model | |
| CN107393126B (en) | A kind of method, apparatus, equipment and the storage medium of bank note version classification | |
| CN107680246B (en) | Method and equipment for positioning curve boundary in paper money pattern | |
| KR20200048256A (en) | Apparatus and method for extracting object information | |
| CN108764225A (en) | Identify the method, apparatus and electronic equipment of horizontally-spliced bank note | |
| CN112133100B (en) | Vehicle detection method based on R-CNN | |
| CN107240184B (en) | Method, device and equipment for identifying versions of plastic coins | |
| CN108074323B (en) | Paper money facing identification method and device thereof |
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 |