CN106846610B - A kind of bank note version recognition methods and device - Google Patents
A kind of bank note version recognition methods and device Download PDFInfo
- Publication number
- CN106846610B CN106846610B CN201710041472.5A CN201710041472A CN106846610B CN 106846610 B CN106846610 B CN 106846610B CN 201710041472 A CN201710041472 A CN 201710041472A CN 106846610 B CN106846610 B CN 106846610B
- Authority
- CN
- China
- Prior art keywords
- row
- version
- image
- column
- crown word
- 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
Landscapes
- Image Analysis (AREA)
- Character Discrimination (AREA)
Abstract
The embodiment of the invention discloses a kind of bank note version recognition methods and devices, this method comprises: obtaining the image of setting regions on bank note, wherein include the crown word number of the bank note in the image of the setting regions；Obtain the boundary of crown word number in the image of the setting regions；Determine that the version of the bank note, the embodiment of the present invention can be improved bank note version recognition accuracy according to the boundary of the crown word number.
Description
Technical field
The present embodiments relate to paper currency detection technology more particularly to a kind of bank note version recognition methods and devices.
Background technique
In order to safeguard that the financial order of society, various countries' paper money recognition technology are being continuously improved.Paper money recognition includes face amount face
Identify to identification and forge or true or paper money, the general identification process of bank note is first to carry out the pretreatment such as slant correction to bank note, is then known
Not Chu bank note version and face amount towards, further according to bank note face amount towards carry out forge or true or paper money identification.The version of bank note is known
Do not occupy critically important status in the identification of bank note, if not identifying the version of bank note first, being just far from being, it is true to carry out to bank note
Puppet identifies.
The prior art generally uses the learning methods such as neural network to identify bank note version, but these methods not only need to instruct
Practice a large amount of sample, and algorithm complexity is high, recognition speed is slow, and is easy to be protected from environmental, and accuracy is lower.
Summary of the invention
The embodiment of the present invention provides a kind of bank note version recognition methods and device, to improve bank note version recognition accuracy.
In a first aspect, the embodiment of the invention provides a kind of bank note version recognition methods, comprising:
Obtain the image of setting regions on bank note, wherein include the prefix of the bank note in the image of the setting regions
Number；
Obtain the boundary of crown word number in the image of the setting regions；
The version of the bank note is determined according to the boundary of the crown word number.
Further, the boundary of crown word number includes: in the image for obtaining the setting regions
Binaryzation is carried out to the image of the setting regions, obtains binary image；
Calculate the row projection and column projection of the binary image, wherein the row is projected as in the binary image
The sum of the gray value of every row pixel, it is described to arrange the sum of the gray value for being projected as each column pixel in the binary image；
The boundary that projection determines crown word number in the image of the setting regions is projected and arranged according to the row.
Further, described to project and arrange the side that projection determines crown word number in the image of the setting regions according to the row
Boundary includes:
Judge in the binary image, is all larger than the with the presence or absence of the row projection of the continuous N row of line number from top to bottom
One threshold value, wherein the line number is the number successively carried out from top to bottom to each row in the binary image, and N is certainly
So number；
In the presence of being judged as, it is determined that in the continuous N row of line number, the smallest row of line number is in the setting regions
The coboundary of corresponding behavior crown word number in image；
Judge in the binary image, the column projection with the presence or absence of row number continuous M column from right to left is all larger than the
One threshold value, wherein the row number is the number successively carried out from left to right to each column in the binary image, and M is certainly
So number；
In the presence of being judged as, it is determined that in the continuous M column of the row number, row number is maximum to be listed in the setting regions
The corresponding right margin for being classified as crown word number in image；
The corresponding line number in the coboundary of the crown word number is added into second threshold, obtains the lower boundary as the crown word number
Corresponding line number.
Further, the boundary according to the crown word number determines that the version of the bank note includes:
Judge whether the lower boundary of the crown word number meets the first preset condition, when be judged as meet when, it is determined that it is described
The version of bank note is first group of version, is otherwise second group of version, wherein first preset condition are as follows: in the binaryzation
In image, the row projection that the corresponding row of the lower boundary counts adjacent I row down is all larger than third threshold value, and I is natural number；
Judge whether the right margin of the crown word number of the bank note of first group of version meets the second preset condition, when being judged as
When meeting, it is determined that the version of the bank note is first version, is otherwise the second edition, wherein second preset condition are as follows:
In the binary image, the sum of column projection that several adjacent J are arranged is turned left greater than the 4th threshold value from the corresponding column of the right margin；
Judge whether the right margin of the crown word number of the bank note of second group of version meets third preset condition, when being judged as
When meeting, it is determined that the version of the bank note is third version, is otherwise fourth edition, wherein the third preset condition are as follows:
In the binary image, by the right margin it is corresponding arrange toward left preset columns after it is corresponding column be denoted as starting column,
The sum of the column projection arranged from the starting to all column the corresponding column of the right margin is greater than the 4th threshold value.
Further, the first version is 2014 versions, and the second edition is 1997 versions, and the third version is
2004 versions, the fourth edition are 1991 versions.
Further, the bank note is Cuba's coin.
Second aspect, the embodiment of the invention also provides a kind of bank note version recognition device, which includes:
Image collection module, for obtaining the image of setting regions on bank note, wherein wrapped in the image of the setting regions
Crown word number containing the bank note；
Boundary obtains module, the boundary of crown word number in the image for obtaining the setting regions；
Version determining module, for determining the version of the bank note according to the boundary of the crown word number.
Further, the boundary acquisition module includes:
Image processing unit carries out binaryzation for the image to the setting regions, obtains binary image；
Computing unit, for calculating the row projection and column projection of the binary image, wherein the row is projected as described
The sum of the gray value of every row pixel in binary image, it is described to arrange the ash for being projected as each column pixel in the binary image
The sum of angle value；
Boundary determination unit determines crown word number in the image of the setting regions for projecting and arranging projection according to the row
Boundary.
Further, the boundary determination unit is specifically used for:
Judge in the binary image, is all larger than the with the presence or absence of the row projection of the continuous N row of line number from top to bottom
One threshold value, wherein the line number is the number successively carried out from top to bottom to each row in the binary image, and N is certainly
So number；
In the presence of being judged as, it is determined that in the continuous N row of line number, the smallest row of line number is in the setting regions
The coboundary of corresponding behavior crown word number in image；
Judge in the binary image, the column projection with the presence or absence of row number continuous M column from right to left is all larger than the
One threshold value, wherein the row number is the number successively carried out from left to right to each column in the binary image, and M is certainly
So number；
In the presence of being judged as, it is determined that in the continuous M column of the row number, row number is maximum to be listed in the setting regions
The corresponding right margin for being classified as crown word number in image；
The corresponding line number in the coboundary of the crown word number is added into second threshold, obtains the lower boundary as the crown word number
Corresponding line number.
Further, the version determining module is specifically used for:
Judge whether the lower boundary of the crown word number meets the first preset condition, when be judged as meet when, it is determined that it is described
The version of bank note is first group of version, is otherwise second group of version, wherein first preset condition are as follows: in the binaryzation
In image, the row projection that the corresponding row of the lower boundary counts adjacent I row down is all larger than third threshold value, and I is natural number；
Judge whether the right margin of the crown word number of the bank note of first group of version meets the second preset condition, when being judged as
When meeting, it is determined that the version of the bank note is first version, is otherwise the second edition, wherein second preset condition are as follows:
In the binary image, the sum of column projection that several adjacent J are arranged is turned left greater than the 4th threshold value from the corresponding column of the right margin；
Judge whether the right margin of the crown word number of the bank note of second group of version meets third preset condition, when being judged as
When meeting, it is determined that the version of the bank note is third version, is otherwise fourth edition, wherein the third preset condition are as follows:
In the binary image, by the right margin it is corresponding arrange toward left preset columns after it is corresponding column be denoted as starting column,
The sum of the column projection arranged from the starting to all column the corresponding column of the right margin is greater than the 4th threshold value.
The image that the embodiment of the present invention passes through setting regions on acquisition bank note, wherein wrapped in the image of the setting regions
Crown word number containing the bank note obtains the boundary of crown word number in the image of the setting regions, according to the boundary of the crown word number
It determines the version of the bank note, can achieve the effect that improve the recognition accuracy of bank note version.
Detailed description of the invention
Fig. 1 is the flow chart of one of the embodiment of the present invention one bank note version recognition methods；
Fig. 2 is the flow chart of one of embodiment of the present invention two bank note version recognition methods；
Fig. 3 is the structural schematic diagram of one of embodiment of the present invention three bank note version recognition device；
Fig. 4 A is bank note schematic diagram in the embodiment of the present invention one；
Fig. 4 B is bank note schematic diagram in the embodiment of the present invention one；
Fig. 4 C is bank note schematic diagram in the embodiment of the present invention one；
Fig. 4 D is bank note schematic diagram in the embodiment of the present invention one；
Fig. 4 E is bank note setting regions schematic diagram in the embodiment of the present invention one；
Fig. 4 F is the boundary schematic diagram of crown word number in the image of setting regions in the embodiment of the present invention one；
Fig. 5 A is binary image schematic diagram in the embodiment of the present invention two；
Fig. 5 B is binary image schematic diagram in the embodiment of the present invention two；
Fig. 5 C is binary image schematic diagram in the embodiment of the present invention two；
Fig. 5 D is binary image schematic diagram in the embodiment of the present invention two；
Fig. 6 A is the boundary schematic diagram of crown word number in the image of setting regions in the embodiment of the present invention two；
Fig. 6 B is the boundary schematic diagram of crown word number in the image of setting regions in the embodiment of the present invention two；
Fig. 6 C is the boundary schematic diagram of crown word number in the image of setting regions in the embodiment of the present invention two；
Fig. 6 D is the boundary schematic diagram of crown word number in the image of setting regions in the embodiment of the present invention two.
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 bank note version recognition methods that the embodiment of the present invention one provides, and the present embodiment is applicable
In bank note version identification the case where, this method can be executed by the bank note version recognition device in the embodiment of the present invention, the dress
Setting can be used the mode of software and/or hardware and realizes, as shown in Figure 1, this method specifically comprises the following steps:
S110 obtains the image of setting regions on bank note, wherein comprising the bank note in the image of the setting regions
Crown word number.
Wherein, the setting regions is to contain the region of crown word number (version flag), the figure of the setting regions on bank note
Picture is the area image comprising crown word number, such as be can be, if the bank note is 5 yuan of 1991 version of Cuba's coin, the setting
Region can be with are as follows: the region of the crown word number comprising " 1991 ".The image of setting regions is the version flag on Cuba's coin comprising 5 yuan
Region image.
Specifically, including the image in the region of the crown word number of bank note on acquisition bank note, the big small capital of setting regions is sent out
Bright embodiment is not limited.
By taking Cuba's coin as an example, as shown in Figure 4 A, setting regions 410 is the region comprising crown word number " 1991 ", such as Fig. 4 B institute
Show, setting regions 420 is the region comprising crown word number " 1997 ", and as shown in Figure 4 C, setting regions 430 is to include crown word number
The region of " 2004 ", as shown in Figure 4 D, setting regions 440 are the region comprising crown word number " 2014 ".It as shown in Figure 4 E, is 2004
The image of setting regions on version Cuba coin comprising crown word number " 2004 ".
S120 obtains the boundary of crown word number in the image of the setting regions.
Wherein, the boundary of the crown word number includes: the right side of the coboundary of crown word number, the lower boundary of crown word number and crown word number
Boundary.
Specifically, can be realized according to the difference of crown word number since the crown word number of the bank note of different editions is different
The identification of bank note version, by taking Cuba's coin of 1991 versions and 2014 versions as an example, 5 yuan of Cuba's coin of crown word number of 1991 versions is
" 1991 ", the crown word numbers of 2014 versions is " 2014 ", as shown in Fig. 4 A and Fig. 4 D, crown word number " 1991 " and crown word number " 2014 "
Font size is different, and the font of the font ratio " 2014 " of " 1991 " is small, therefore can be according to the boundary of crown word number to the version of bank note
Originally it distinguishes.Similarly, Cuba's coin of 1997 versions and 2004 versions can also be distinguished.
By taking Cuba's coin of 2004 versions as an example, as illustrated in figure 4f, the boundary of crown word number includes: coboundary 460, lower boundary
470 and right margin 450.
S130 determines the version of the bank note according to the boundary of the crown word number.
Specifically, according to the difference on the boundary of the crown word number and the positional relationship of crown word number, and then distinguish the version of bank note
This.
Optionally, the bank note is Cuba's coin.
Optionally, the bank note version can be 2014 versions, or 1997 versions can also be 2004 versions,
It can also be 1991 versions.
The technical solution of the present embodiment, by the image for obtaining setting regions on bank note, wherein the figure of the setting regions
Comprising the crown word number of the bank note as in, the boundary of crown word number in the image of the setting regions is obtained, according to the crown word number
Boundary determine the version of the bank note, can achieve the effect that improve the recognition accuracy of bank note version.
Embodiment two
Fig. 2 is the flow chart of one of the embodiment of the present invention two bank note version recognition methods, and the present embodiment is with above-mentioned reality
It applies and optimizes based on example, in the present embodiment, the boundary of crown word number includes: in the image for obtaining the setting regions
Binaryzation is carried out to the image of the setting regions, obtains binary image；Calculate row projection and the column of the binary image
Projection, wherein the row is projected as the sum of the gray value of every row pixel in the binary image, and the column are projected as described
The sum of the gray value of each column pixel in binary image；The figure that projection determines the setting regions is projected and arranged according to the row
The boundary of crown word number as in.
Binary conversion treatment is carried out by the image to predeterminable area as a result, can reduce the image progress to setting regions
The complexity that row projection and column projection calculate, and then achieve the effect that improve paper money recognition speed.
As shown in Fig. 2, the method for the present embodiment specifically comprises the following steps:
S210 obtains the image of setting regions on bank note, wherein comprising the bank note in the image of the setting regions
Crown word number.
Specifically, by taking 5 yuan of 2004 version of Cuba's coin as an example, obtain the image of setting regions on bank note, setting regions it is big
Small no strict regulations, but must include 5 yuan of Cuba's coin of version flag, that is, crown word number.For example, as shown in Figure 4 C, obtaining
The image of setting regions 410 on 5 yuan of 2004 version bank note of Cuba's coin, image such as Fig. 4 E institute of setting regions on the bank note of acquisition
Show.
S220 carries out binaryzation to the image of the setting regions, obtains binary image.
Wherein, the binaryzation of described image exactly sets 0 or 1 for the gray value of the pixel on image, that is, will
Whole image, which shows, significantly only has black and white visual effect.The binary image is to carry out to the image of setting regions
The image that binary conversion treatment obtains.The two-dimensional matrix of binary image is only made of 0,1 two value, and " 0 " represents black, " 1 " generation
White.Since value only has 0,1 two kind of possibility to each pixel (each element in matrix), so binary image in computer
Data type is usually 1 binary digit.It as shown in Figure 5A, is the binaryzation of setting regions image in Cuba's coin of 1991 versions
Image, the gray value of the pixel of the black portions in figure are 0, and the gray value of the pixel of the white portion in figure is 1.
S230 calculates the row projection and column projection of the binary image, wherein the row is projected as the binary picture
As in every row pixel the sum of gray value, it is described arrange be projected as each column pixel in the binary image gray value it
With.
Wherein, the gray value of the binary image is with raw column data matrix representation.It calculates and described has carried out binary conversion treatment
Setting regions image row projection and column projection, it may be assumed that calculate every row pixel of binary image whole rows for including
The sum of gray value and the sum of the gray value for calculating each column pixel all arranged that binary image includes.
S240 projects and arranges the boundary that projection determines crown word number in the image of the setting regions according to the row.
Wherein, the mode for obtaining the boundary of crown word number in the image of the setting regions can be using first to setting regions
Image carries out binary conversion treatment, then calculates the row projection and column projection of binary image, and then obtains the image of setting regions
The boundary of middle crown word number, wherein the row is projected as the sum of the gray value of every row pixel in the binary image, the column
It is projected as the sum of the gray value of each column pixel in the binary image；Obtain crown word number in the image of the setting regions
The mode on boundary can also be using the row projection and column projection for the image for directly calculating setting regions, and then obtains setting regions
The boundary of crown word number in image.
Optionally, described to project and arrange the boundary that projection determines crown word number in the image of the setting regions according to the row
Include:
Judge in the binary image, is all larger than the with the presence or absence of the row projection of the continuous N row of line number from top to bottom
One threshold value, wherein the line number is the number successively carried out from top to bottom to each row in the binary image, and N is certainly
So number；
In the presence of being judged as, it is determined that in the continuous N row of line number, the smallest row of line number is in the setting regions
The coboundary of corresponding behavior crown word number in image；
Judge in the binary image, the column projection with the presence or absence of row number continuous M column from right to left is all larger than the
One threshold value, wherein the row number is the number successively carried out from left to right to each column in the binary image, and M is certainly
So number；
In the presence of being judged as, it is determined that in the continuous M column of the row number, row number is maximum to be listed in the setting regions
The corresponding right margin for being classified as crown word number in image；
The corresponding line number in the coboundary of the crown word number is added into second threshold, obtains the lower boundary as the crown word number
Corresponding line number.
By taking Cuba's coin as an example, if binary image from top to bottom to each traveling row number, is followed successively by the 1st row, the 2nd
Row ..., the 30th row, there are the row projections of the 4th row to be greater than the row projection of the 2 and the 5th row also greater than 2, it is determined that the 4th behavior crown word number
Coboundary；
If binary image is from left to right numbered each column, the 1st column are followed successively by, the 2nd column ..., the 100th arranges, from
Right-to-left calculates column projection, and the column for being greater than the 2 and the 40th column there are the column projection of the 41st column are projected also greater than 2, it is determined that
41st is classified as the right margin of crown word number；
The second threshold is the minimum value of the height of 5 yuan of Cuba's coin of prefix sign character of different editions, in this implementation
In example, the height minimum 8 of 5 yuan of Cuba's coin of prefix sign character of 1991 versions, then the corresponding line number of the lower boundary of crown word number
Line number corresponding equal to coboundary adds 8, such as can be, if the coboundary of crown word number is the 4th row, the lower boundary of crown word number
For the 12nd row.
S250 determines the version of the bank note according to the boundary of the crown word number.
By taking Cuba's coin as an example, as shown in Fig. 6 A, 6B, 6C and 6D, it can be seen that 5 yuan of 1991 version of Cuba's coin, 1997 versions,
The boundary of the crown word number of 2004 versions and 2014 versions and the positional relationship of crown word number be not identical, therefore can be according to crown word number
Boundary determine the version of bank note.For example, the crown word number boundary of 5 yuan of 2004 version of 5 yuan of 1991 version of Cuba's coin and Cuba's coin is several
Entire crown word number is integrally incorporated in the boundary, 5 yuan of 1997 version of Cuba's coin, the crown word number of 2014 version of Cuba's coin
Boundary is there is no crown word number to be included in the boundary, and a part of crown word number is in the outside on the crown word number boundary, therefore,
5 yuan of Cuba's coin of version can be divided into two groups according to features described above, 5 yuan of 1991 version of Cuba's coin and 5 yuan of Cuba's coin
2004 versions are one group, and 5 yuan of 2014 version of 5 yuan of 1997 version of Cuba's coin and Cuba's coin is one group.Due to Cuba's coin 5 yuan 1991 editions
This is different from the last one number of crown word number of Cuba's 5 yuan of 2004 version of coin, therefore arranges projection difference, and then can be according to upper
State 1991 versions and 2004 versions that difference distinguishes 5 yuan of Cuba's coin.Similarly, 5 yuan of 1997 version of Cuba's coin and 2014 versions are distinguished.
Optionally, the boundary according to the crown word number determines that the version of the bank note includes:
Judge whether the lower boundary of the crown word number meets the first preset condition, when be judged as meet when, it is determined that it is described
The version of bank note is first group of version, is otherwise second group of version, wherein first preset condition are as follows: in the binaryzation
In image, the row projection that the corresponding row of the lower boundary counts adjacent I row down is all larger than third threshold value, and I is natural number；
Judge whether the right margin of the crown word number of the bank note of first group of version meets the second preset condition, when being judged as
When meeting, it is determined that the version of the bank note is first version, is otherwise the second edition, wherein second preset condition are as follows:
In the binary image, the sum of column projection that several adjacent J are arranged is turned left greater than the 4th threshold value from the corresponding column of the right margin；
Judge whether the right margin of the crown word number of the bank note of second group of version meets third preset condition, when being judged as
When meeting, it is determined that the version of the bank note is third version, is otherwise fourth edition, wherein the third preset condition are as follows:
In the binary image, by the right margin it is corresponding arrange toward left preset columns after it is corresponding column be denoted as starting column,
The sum of the column projection arranged from the starting to all column the corresponding column of the right margin is greater than the 4th threshold value.
Still by taking Cuba's coin as an example, if the corresponding line number of lower boundary be the 12nd row, if the 13rd row, the 14th row, the 15th row with
And the 16th row row projection both greater than 3, then it is assumed that be first group of version.Wherein, first group of version include 1997 versions and
2014 versions.The second edition includes 1991 versions and 2004 versions；
If the corresponding row number of right margin is the 40th column, if the sum of the column projection of the 39th column, the 38th column and the 37th column is greater than
20, it is determined that the bank note version is 2014 versions.Conversely, being then considered 1997 versions；
If the corresponding row number of right margin is the 40th column, right margin is denoted as starting column toward the column of left 2, if the 38th column, the
The sum of 39 column and the column projection of the 40th column are greater than 20, it is determined that the bank note version is 2004 versions.Conversely, being then considered 1991
Version.
Wherein, the N row can be 2 rows, and the first threshold can be 2, and the M column can be 2 column, second threshold
Value can be 8, and the I row can be 4 rows, and the third threshold value can be 3, and the J column can be 3 column, the 4th threshold value
It can be 20, the default columns can be 2.
Optionally, the first version is 2014 versions, and the second edition is 1997 versions, and the third version is
2004 versions, the fourth edition are 1991 versions.
In a specific example, as shown in Fig. 4 A, 4B, 4C and 4D, just to 5 yuan of Cuba's coin of several versions of fronts
To setting regions 410, setting regions 420, setting regions 430 and setting regions 440 intercepted.Truncated picture is carried out
Binaryzation obtains binary image, that is, the case where the gray value of pixel is reduced to only 0 or 1, as Fig. 5 A, 5B,
Shown in 5C and 5D.Find coboundary and the right margin of crown word number.Row projection column projection is asked to the image after binaryzation, from top to bottom
If continuous 2 rows projection is greater than 2 (empirical values), then it is assumed that the minimum row of this two row is the coboundary of crown word number, is turned left from the right side,
If continuous two column projection is greater than 2, the right margin of crown word number will be determined as in the maximum column of this two column.Base in coboundary
It is determined as under the crown word number of all versions in the present embodiment on plinth plus 8 rows (minimum value of the height of prefix sign character)
Boundary.The boundary of the crown word number is the boundary of estimation, is not true boundary.Lower boundary down 4 rows row projection it is all big
In 3, it is determined that for 1997 versions either 2014 versions, be otherwise determined as 1991 versions and 2004 versions.It has been divided into two in this way
Major class.1st class is 1997 and 2014 versions, and the 2nd class is 1991 and 2001 versions.For 1991 and 2004, basis on the right
It is upper to be denoted as starting column toward the column of left 2 (empirical value), the sums that this starting column arrives all column projections of right margin are calculated, and if big
Then it is determined as 2004 versions in 20 (empirical values).Otherwise it is determined as 1991 versions.For 1997 and 2014, calculates right margin and turn left
The sum of 3 column are greater than 20, it is determined that are 2014 versions, otherwise are determined as 1997 versions.
The technical solution of the present embodiment carries out binaryzation by the image to the setting regions, obtains binary image,
The row projection and column projection of binary image are calculated, row is projected as the sum of the gray value of every row pixel in binary image, column
It is projected as the sum of the gray value of each column pixel in binary image, is projected according to row and column projects the image for determining setting regions
The gray value of the pixel of the image of setting regions is reduced to 0 or 1 by binaryzation, by upper by the boundary of middle crown word number
The method of stating, which can reduce, carries out the complexity that capable projection and column projection calculate to the image of setting regions, and then reaches raising paper
The effect of coin recognition speed.
Embodiment three
Fig. 3 is a kind of structural schematic diagram of bank note version recognition device of the embodiment of the present invention three.The present embodiment is applicable
In bank note version identification the case where, the mode which can be used software and/or hardware is realized, which can be integrated in any mention
In the equipment of bank note version identification, as shown in figure 3, the bank note version recognition device specifically includes: image collection module 310,
Boundary obtains module 320 and version determining module 330.
Wherein, image collection module 310, for obtaining the image of setting regions on bank note, wherein the setting regions
It include the crown word number of the bank note in image；
Boundary obtains module 320, the boundary of crown word number in the image for obtaining the setting regions；
Version determining module 330, for determining the version of the bank note according to the boundary of the crown word number.
Optionally, the boundary acquisition module 320 includes:
Image processing unit carries out binaryzation for the image to the setting regions, obtains binary image；
Computing unit, for calculating the row projection and column projection of the binary image, wherein the row is projected as described
The sum of the gray value of every row pixel in binary image, it is described to arrange the ash for being projected as each column pixel in the binary image
The sum of angle value；
Boundary determination unit determines crown word number in the image of the setting regions for projecting and arranging projection according to the row
Boundary.
Optionally, the boundary determination unit is specifically used for:
Judge in the binary image, is all larger than the with the presence or absence of the row projection of the continuous N row of line number from top to bottom
One threshold value, wherein the line number is the number successively carried out from top to bottom to each row in the binary image, and N is certainly
So number；
In the presence of being judged as, it is determined that in the continuous N row of line number, the smallest row of line number is in the setting regions
The coboundary of corresponding behavior crown word number in image；
Judge in the binary image, the column projection with the presence or absence of row number continuous M column from right to left is all larger than the
One threshold value, wherein the row number is the number successively carried out from left to right to each column in the binary image, and M is certainly
So number；
In the presence of being judged as, it is determined that in the continuous M column of the row number, row number is maximum to be listed in the setting regions
The corresponding right margin for being classified as crown word number in image；
The corresponding line number in the coboundary of the crown word number is added into second threshold, obtains the lower boundary as the crown word number
Corresponding line number.
Optionally, the version determining module 330 is specifically used for:
Judge whether the lower boundary of the crown word number meets the first preset condition, when be judged as meet when, it is determined that it is described
The version of bank note is first group of version, is otherwise second group of version, wherein first preset condition are as follows: in the binaryzation
In image, the row projection that the corresponding row of the lower boundary counts adjacent I row down is all larger than third threshold value, and I is natural number；
Judge whether the right margin of the crown word number of the bank note of first group of version meets the second preset condition, when being judged as
When meeting, it is determined that the version of the bank note is first version, is otherwise the second edition, wherein second preset condition are as follows:
In the binary image, the sum of column projection that several adjacent J are arranged is turned left greater than the 4th threshold value from the corresponding column of the right margin；
Judge whether the right margin of the crown word number of the bank note of second group of version meets third preset condition, when being judged as
When meeting, it is determined that the version of the bank note is third version, is otherwise fourth edition, wherein the third preset condition are as follows:
In the binary image, by the right margin it is corresponding arrange toward left preset columns after it is corresponding column be denoted as starting column,
The sum of the column projection arranged from the starting to all column the corresponding column of the right margin is greater than the 4th threshold value.
The technical solution of the present embodiment, by the image for obtaining setting regions on bank note, wherein the figure of the setting regions
Comprising the crown word number of the bank note as in, the boundary of crown word number in the image of the setting regions is obtained, according to the crown word number
Boundary determine the version of the bank note, can achieve the effect that improve the recognition accuracy of bank note version.
Method provided by any embodiment of the invention can be performed in the said goods, has the corresponding functional module of execution method
And beneficial effect.
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 (4)
1. a kind of bank note version recognition methods characterized by comprising
Obtain the image of setting regions on bank note, wherein include the crown word number of the bank note in the image of the setting regions；
Obtain the boundary of crown word number in the image of the setting regions；
The version of the bank note is determined according to the boundary of the crown word number；
Wherein, the boundary of crown word number includes: in the image for obtaining the setting regions
Binaryzation is carried out to the image of the setting regions, obtains binary image；
Calculate the row projection and column projection of the binary image, wherein the row is projected as every row in the binary image
The sum of gray value of pixel, it is described to arrange the sum of the gray value for being projected as each column pixel in the binary image；
The boundary that projection determines crown word number in the image of the setting regions is projected and arranged according to the row；
Wherein, the boundary for projecting and arranging crown word number in the image of the determining setting regions of projection according to the row includes:
Judge in the binary image, is all larger than the first threshold with the presence or absence of the row projection of the continuous N row of line number from top to bottom
Value, wherein the line number is the number successively carried out from top to bottom to each row in the binary image, and N is natural number；
In the presence of being judged as, it is determined that in the continuous N row of line number, image of the smallest row of line number in the setting regions
In corresponding behavior crown word number coboundary；
Judge in the binary image, the column projection with the presence or absence of the continuous M column of row number is all larger than the first threshold from right to left
Value, wherein the row number is the number successively carried out from left to right to each column in the binary image, and M is natural number；
In the presence of being judged as, it is determined that in the continuous M column of row number, the maximum image for being listed in the setting regions of row number
In the corresponding right margin for being classified as crown word number；
The corresponding line number in the coboundary of the crown word number is added into second threshold, is obtained corresponding as the lower boundary of the crown word number
Line number；
Wherein, the boundary according to the crown word number determines that the version of the bank note includes:
Judge whether the lower boundary of the crown word number meets the first preset condition, when be judged as meet when, it is determined that the bank note
Version be first group of version, be otherwise second group of version, wherein first preset condition are as follows: in the binary image
In, the row projection that the corresponding row of the lower boundary counts adjacent I row down is all larger than third threshold value, and I is natural number；
Judge whether the right margin of the crown word number of the bank note of first group of version meets the second preset condition, meets when being judged as
When, it is determined that the version of the bank note is first version, is otherwise the second edition, wherein second preset condition are as follows: in institute
It states in binary image, turns left the sum of column projection that several adjacent J are arranged greater than the 4th threshold value from the corresponding column of the right margin；
Judge whether the right margin of the crown word number of the bank note of second group of version meets third preset condition, meets when being judged as
When, it is determined that the version of the bank note is third version, is otherwise fourth edition, wherein the third preset condition are as follows: in institute
State in binary image, by the right margin it is corresponding arrange toward left preset columns after it is corresponding column be denoted as starting column, from institute
The sum of the column projection for stating starting column to all column between the corresponding column of the right margin is greater than the 4th threshold value.
2. the method according to claim 1, wherein the first version is 2014 versions, the second edition
For 1997 versions, the third version is 2004 versions, and the fourth edition is 1991 versions.
3. method described in any one of -2 according to claim 1, which is characterized in that the bank note is Cuba's coin.
4. a kind of bank note version recognition device characterized by comprising
Image collection module, for obtaining the image of setting regions on bank note, wherein include institute in the image of the setting regions
State the crown word number of bank note；
Boundary obtains module, the boundary of crown word number in the image for obtaining the setting regions；
Version determining module, for determining the version of the bank note according to the boundary of the crown word number；
Wherein, the boundary acquisition module includes:
Image processing unit carries out binaryzation for the image to the setting regions, obtains binary image；
Computing unit, for calculating the row projection and column projection of the binary image, wherein the row is projected as the two-value
Change the sum of the gray value of every row pixel in image, it is described to arrange the gray value for being projected as each column pixel in the binary image
The sum of；
Boundary determination unit, for projecting and arranging the side that projection determines crown word number in the image of the setting regions according to the row
Boundary；
Wherein, the boundary determination unit is specifically used for:
Judge in the binary image, is all larger than the first threshold with the presence or absence of the row projection of the continuous N row of line number from top to bottom
Value, wherein the line number is the number successively carried out from top to bottom to each row in the binary image, and N is natural number；
In the presence of being judged as, it is determined that in the continuous N row of line number, image of the smallest row of line number in the setting regions
In corresponding behavior crown word number coboundary；
Judge in the binary image, the column projection with the presence or absence of the continuous M column of row number is all larger than the first threshold from right to left
Value, wherein the row number is the number successively carried out from left to right to each column in the binary image, and M is natural number；
In the presence of being judged as, it is determined that in the continuous M column of row number, the maximum image for being listed in the setting regions of row number
In the corresponding right margin for being classified as crown word number；
The corresponding line number in the coboundary of the crown word number is added into second threshold, is obtained corresponding as the lower boundary of the crown word number
Line number；
Wherein, the version determining module is specifically used for:
Judge whether the lower boundary of the crown word number meets the first preset condition, when be judged as meet when, it is determined that the bank note
Version be first group of version, be otherwise second group of version, wherein first preset condition are as follows: in the binary image
In, the row projection that the corresponding row of the lower boundary counts adjacent I row down is all larger than third threshold value, and I is natural number；
Judge whether the right margin of the crown word number of the bank note of first group of version meets the second preset condition, meets when being judged as
When, it is determined that the version of the bank note is first version, is otherwise the second edition, wherein second preset condition are as follows: in institute
It states in binary image, turns left the sum of column projection that several adjacent J are arranged greater than the 4th threshold value from the corresponding column of the right margin；
Judge whether the right margin of the crown word number of the bank note of second group of version meets third preset condition, meets when being judged as
When, it is determined that the version of the bank note is third version, is otherwise fourth edition, wherein the third preset condition are as follows: in institute
State in binary image, by the right margin it is corresponding arrange toward left preset columns after it is corresponding column be denoted as starting column, from institute
The sum of the column projection for stating starting column to all column between the corresponding column of the right margin is greater than the 4th threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710041472.5A CN106846610B (en) | 2017-01-20 | 2017-01-20 | A kind of bank note version recognition methods and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710041472.5A CN106846610B (en) | 2017-01-20 | 2017-01-20 | A kind of bank note version recognition methods and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106846610A CN106846610A (en) | 2017-06-13 |
| CN106846610B true CN106846610B (en) | 2019-04-26 |
Family
ID=59119313
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710041472.5A Active CN106846610B (en) | 2017-01-20 | 2017-01-20 | A kind of bank note version recognition methods and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106846610B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107657712A (en) * | 2017-09-08 | 2018-02-02 | 深圳怡化电脑股份有限公司 | A kind of paper currency detecting method and note detection device |
| CN108388870A (en) * | 2018-02-28 | 2018-08-10 | 北京奇艺世纪科技有限公司 | A kind of picture material repeatability detection method and device |
| CN111914717A (en) * | 2020-07-24 | 2020-11-10 | 安徽华速达电子科技有限公司 | Data entry method and device based on meter reading data intelligent identification |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101751568A (en) * | 2008-12-12 | 2010-06-23 | 汉王科技股份有限公司 | ID No. locating and recognizing method |
| JP2011113348A (en) * | 2009-11-27 | 2011-06-09 | Musashi Eng Co Ltd | Device and method for detecting number of bundles |
| CN102722711A (en) * | 2012-05-31 | 2012-10-10 | 信帧电子技术（北京）有限公司 | License plate character sequence identification method and device in license plate type identification |
| CN103093181A (en) * | 2011-11-01 | 2013-05-08 | 青岛海信网络科技股份有限公司 | License plate image locating method and device |
| CN103870803A (en) * | 2013-10-21 | 2014-06-18 | 北京邮电大学 | Vehicle license plate recognition method and system based on coarse positioning and fine positioning fusion |
| CN104318238A (en) * | 2014-11-10 | 2015-01-28 | 广州御银科技股份有限公司 | Method for extracting crown word numbers from scanned banknote images in banknote detection module |
| CN105069900A (en) * | 2015-08-14 | 2015-11-18 | 深圳怡化电脑股份有限公司 | Method and device for processing banknote information |
| CN106296975A (en) * | 2016-08-05 | 2017-01-04 | 深圳怡化电脑股份有限公司 | The recognition methods of a kind of dollar bill face amount and device |
-
2017
- 2017-01-20 CN CN201710041472.5A patent/CN106846610B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101751568A (en) * | 2008-12-12 | 2010-06-23 | 汉王科技股份有限公司 | ID No. locating and recognizing method |
| JP2011113348A (en) * | 2009-11-27 | 2011-06-09 | Musashi Eng Co Ltd | Device and method for detecting number of bundles |
| CN103093181A (en) * | 2011-11-01 | 2013-05-08 | 青岛海信网络科技股份有限公司 | License plate image locating method and device |
| CN102722711A (en) * | 2012-05-31 | 2012-10-10 | 信帧电子技术（北京）有限公司 | License plate character sequence identification method and device in license plate type identification |
| CN103870803A (en) * | 2013-10-21 | 2014-06-18 | 北京邮电大学 | Vehicle license plate recognition method and system based on coarse positioning and fine positioning fusion |
| CN104318238A (en) * | 2014-11-10 | 2015-01-28 | 广州御银科技股份有限公司 | Method for extracting crown word numbers from scanned banknote images in banknote detection module |
| CN105069900A (en) * | 2015-08-14 | 2015-11-18 | 深圳怡化电脑股份有限公司 | Method and device for processing banknote information |
| CN106296975A (en) * | 2016-08-05 | 2017-01-04 | 深圳怡化电脑股份有限公司 | The recognition methods of a kind of dollar bill face amount and device |
Non-Patent Citations (2)
| Title |
|---|
| 《基于FPGA的车牌识别系统研究与实现》;胡玉琛等;《电子测量技术》;20151231;第38卷(第12期);第57-58页第4.2节第2-3部分 |
| 《基于印刷年份的人民币版本识别技术研究》;罗帅等;《现代电子技术》;20150930;第38卷(第18期);第72-74页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106846610A (en) | 2017-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110046529B (en) | Two-dimensional code identification method, device and equipment | |
| Samra et al. | Localization of license plate number using dynamic image processing techniques and genetic algorithms | |
| US8218875B2 (en) | Method and system for preprocessing an image for optical character recognition | |
| CN106874909A (en) | A kind of recognition methods of image character and its device | |
| CN111178355B (en) | Seal identification method, device and storage medium | |
| CN106846610B (en) | A kind of bank note version recognition methods and device | |
| CN113486828B (en) | Image processing method, device, equipment and storage medium | |
| CN103714327A (en) | Method and system for correcting image direction | |
| CN111461126B (en) | Space recognition method and device in text line, electronic equipment and storage medium | |
| JP2018524663A (en) | Fingerprint ridge point recognition method and apparatus | |
| EP2569930A2 (en) | Segmentation of a word bitmap into individual characters or glyphs during an ocr process | |
| CN108734161B (en) | Method, device and equipment for identifying prefix number area and storage medium | |
| CN110428414B (en) | Method and device for identifying number of notes in image | |
| CN110598566A (en) | Image processing method, device, terminal and computer readable storage medium | |
| CN107240185B (en) | A kind of crown word number identification method, device, equipment and storage medium | |
| CN109389110B (en) | Region determination method and device | |
| CN105184294B (en) | It is a kind of based on pixel tracking inclination text judge recognition methods | |
| JP2019528543A (en) | Digital object unique identifier (DOI) recognition method and apparatus | |
| CN113033543A (en) | Curved text recognition method, device, equipment and medium | |
| CN112926564A (en) | Picture analysis method, system, computer device and computer-readable storage medium | |
| CN106991753A (en) | A kind of image binaryzation method and device | |
| KR101725501B1 (en) | Method and apparatus for recognizing character | |
| CN110533020B (en) | Character information identification method and device and storage medium | |
| CN106951902B (en) | Image binarization processing method and device | |
| CN112749696A (en) | Text detection method and device |
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 |