CN105184955A - Method and device for identifying paper money - Google Patents
Method and device for identifying paper money Download PDFInfo
- Publication number
- CN105184955A CN105184955A CN201510506975.6A CN201510506975A CN105184955A CN 105184955 A CN105184955 A CN 105184955A CN 201510506975 A CN201510506975 A CN 201510506975A CN 105184955 A CN105184955 A CN 105184955A
- Authority
- CN
- China
- Prior art keywords
- safety line
- hollow out
- out safety
- light
- projection value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention discloses a method and a device for identifying paper money. The method comprises: through the position of an optically variable hollowed-out security line of paper money, performing binaryzation on the position of the optically variable hollowed-out security line; calculating a row-projection value of the binary optically variable hollowed-out security line position, according to the row-projection value, obtaining a row-projection value curve; using a dynamic time warping algorithm to match the row-projection value curve; and according to a matching result, identifying authenticity of the optically variable hollowed-out security line. The binaryzation is performed on the optically variable hollowed-out security line position, the row-projection value of the optically variable hollowed-out security line position is obtained, and the projection value curve is obtained. The dynamic time warping algorithm is used to match the row-projection value curve of the optically variable hollowed-out security line. The method can effectively identify authenticity of the optically variable hollowed-out security line of paper money. The device and the method are beneficial for authenticity prevention of the optically variable hollowed-out security lines.
Description
Technical field
The present invention relates to paper money recognition field, particularly relate to a kind of recognition methods and device of bank note.
Background technology
2015 millboard coin are more than existing version bank note, and a light becomes hollow out safety line, and this light becomes hollow out safety line midamble and can clearly in infrared transmission image see.By judging that light becomes the feature of hollow out safety line, can judge whether the version of bank note belongs to 2015 editions, differentiating that light becomes the hollow out safety line true and false simultaneously.
Summary of the invention
The invention provides recognition methods and device that bank note light becomes hollow out safety line, the detection that the method and device become hollow out safety line to the light on bank note is adaptable, can judge that whether the version of bank note is have the version that light becomes hollow out safety line, differentiate that light becomes the true and false of hollow out safety line simultaneously.
For realizing above-mentioned design, the present invention by the following technical solutions:
Have employed a kind of recognition methods of bank note on the one hand, comprising:
The light obtaining bank note becomes hollow out safety line position, becomes hollow out safety line position carry out binaryzation to described light;
Calculate the row projection value that the light after binaryzation becomes hollow out safety line position, obtain row projection value curve according to described row projection value;
Use row projection value curve described in dynamic time warping algorithmic match;
According to matching result, differentiate that light becomes the true and false of hollow out safety line.
Wherein, described acquisition light becomes hollow out safety line position, becomes hollow out safety line position and carries out binaryzation, comprising described light:
Obtain the infrared transmission image of bank note, choose light according to described infrared transmission image and become hollow out safety line position, cut the region subgraph that described light becomes hollow out safety line position;
Calculate the row projection value of described region subgraph, extract according to described row projection value the row projection value feature that light becomes hollow out safety line position;
Accurately locate light according to described row projection value feature and become hollow out safety line position;
Use Adaptive Thresholding to become hollow out safety line position to described light and carry out binaryzation.
Wherein, the row projection value of described calculating described region subgraph, extracts after light becomes the row projection value feature of hollow out safety line position according to described row projection value, describedly accurately locates before light becomes hollow out safety line position according to the described projection value feature of row, also comprises:
Identifying bank note according to the row projection value feature of described light change hollow out safety line position is have the bank note that light becomes hollow out safety line version.
Wherein, the light of described acquisition bank note becomes hollow out safety line position, before carrying out binaryzation, also comprises described light change hollow out safety line position:
Identify the Information of bank note, described Information is divided into: front forward, front oppositely, reverse side forward and reverse side reverse;
The light of described acquisition bank note becomes hollow out safety line position, becomes hollow out safety line position and carries out binaryzation, be specially described light:
When the Information identifying bank note is front forward or reverse side is reverse, the light obtaining bank note from the right side of bank note becomes hollow out safety line position, becomes hollow out safety line position carry out binaryzation to described light;
When the Information identifying bank note be front oppositely or reverse side forward time, the light obtaining bank note from the left side of bank note becomes hollow out safety line position, becomes hollow out safety line position carry out binaryzation to described light.
Wherein, row projection value curve described in described use dynamic time warping algorithmic match, is specially:
Find a best Time alignment function, make described row projection value curve closest to standard row projection value curve, obtain the minor increment summation of match point time.
Wherein, described according to matching result, differentiate that light becomes the true and false of hollow out safety line, be specially:
If the minor increment summation of described match point time is within the scope of preset distance, then described light becomes hollow out safety line into true, otherwise is false.
Adopt a kind of recognition device of bank note on the other hand, comprising:
Acquiring unit, becomes hollow out safety line position for the light obtaining bank note, becomes hollow out safety line position carry out binaryzation to described light;
Row projecting cell, becomes the row projection value of hollow out safety line position for calculating the light after binaryzation, obtain row projection value curve according to described row projection value;
Matching unit, for using row projection value curve described in dynamic time warping algorithmic match;
False distinguishing unit, for according to matching result, differentiates that light becomes the true and false of hollow out safety line.
Wherein, also comprise:
Towards recognition unit, for identifying the Information of bank note, described Information is divided into: front forward, front oppositely, reverse side forward and reverse side reverse.
Wherein, also comprise:
Version recognition unit, it is have the bank note that light becomes hollow out safety line version that the row projection value feature for becoming hollow out safety line position according to described light identifies bank note.
Wherein, described acquiring unit comprises:
Cutter unit, for obtaining the infrared transmission image of bank note, choosing light according to described infrared transmission image and becoming hollow out safety line position, cut the region subgraph that described light becomes hollow out safety line position;
Row projecting cell, for calculating the row projection value of described region subgraph, extracts according to described row projection value the row projection value feature that light becomes hollow out safety line position;
Positioning unit, becomes hollow out safety line position for accurately locating light according to described row projection value feature;
Binarization unit, becomes hollow out safety line position for using Adaptive Thresholding to described light and carries out binaryzation.
Beneficial effect of the present invention is: the present invention becomes hollow out safety line position by the light obtaining bank note, becomes hollow out safety line position carry out binaryzation to described light; Calculate the row projection value that the light after binaryzation becomes hollow out safety line position, obtain row projection value curve according to described row projection value; Use row projection value curve described in dynamic time warping algorithmic match; According to matching result, differentiate that light becomes the true and false of hollow out safety line.The present invention carries out binaryzation to light change hollow out safety line position and obtains the row projection value of light change hollow out safety line position and obtain row projection value curve, dynamic time warping algorithm is used to mate the row projection value curve that light becomes hollow out safety line, effectively can become hollow out safety line to the light of bank note and carry out false distinguishing identification, be conducive to the true and false prevention that light becomes hollow out safety line.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing the embodiment of the present invention is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to the content of the embodiment of the present invention and these accompanying drawings.
Fig. 1 is the method flow diagram of the first embodiment of the recognition methods of a kind of bank note provided in the specific embodiment of the invention.
Fig. 2 is the method flow diagram of the second embodiment of the recognition methods of a kind of bank note provided in the specific embodiment of the invention.
Fig. 3 is the bank note schematic diagram of the miscellaneous editions provided in the specific embodiment of the invention.
Fig. 4 is the region subgraph row projection value curve map of the light change hollow out safety line position of the bank note of miscellaneous editions in Fig. 3.
Fig. 5 is the binary image of the light change hollow out safety line position of bank note in Fig. 3-c.
Fig. 6 is the row projection value curve after Fig. 5 binaryzation.
Fig. 7 is the block diagram of the first embodiment of the recognition device of a kind of bank note provided in the specific embodiment of the invention.
Fig. 8 is the block diagram of the second embodiment of the recognition device of a kind of bank note provided in the specific embodiment of the invention.
Embodiment
The technical matters solved for making the present invention, the technical scheme of employing and the technique effect that reaches are clearly, be described in further detail below in conjunction with the technical scheme of accompanying drawing to the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those skilled in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Please refer to Fig. 1, it is the method flow diagram of the first embodiment of the recognition methods of a kind of bank note provided in the specific embodiment of the invention.As shown in the figure, this recognition methods, comprising:
Step S101, the light obtaining bank note becomes hollow out safety line position, becomes hollow out safety line position carry out binaryzation to described light.
Concrete, the word in the middle of hollow out safety line can be become by display light clearly by infrared transmission image, adopt Adaptive Thresholding to become hollow out safety line position to light according to word in the distribution that light becomes in hollow out safety line and carry out binaryzation.Wherein, Adaptive Thresholding is that whole region light being become hollow out safety line position is divided into three parts by word: the left side, centre, the right, and middle part is the region of word, tries to achieve the mean value of the left side and the right gray scale as gray threshold.In binarization, being black when gray-scale value is more than or equal to gray threshold, is white when gray-scale value is less than gray threshold, thus completes the binaryzation that light becomes hollow out safety line position, effectively can carry out denoising to background, greatly simplify the complexity of calculating.
Step S102, calculates the row projection value that the light after binaryzation becomes hollow out safety line position, obtains row projection value curve according to described row projection value.
Wherein, bank note infrared view width is 200DPI resolution, be highly 150DPI resolution, the gray-scale value sum that light becomes every a line in hollow out safety line forms a row projection value, the row projection value that whole light becomes hollow out safety line position forms row projection value curve, after binaryzation, projection value can be larger when running into font for safety line row projection value curve, when not having font, projection value can be less, projection value curve be with font with or without and font change and change, be illustrated in figure 5 the image after the bank note light change hollow out safety line binaryzation of 2015 editions 100 yuan of face amounts, row drop shadow curve as corresponding in Fig. 6.
Step S103, uses row projection value curve described in dynamic time warping algorithmic match.
Wherein, dynamic time warping algorithm (DynamicTimeWarping, DTW) thought is: because voice signal is a kind of signal with quite large randomness, even if identical speaker is to identical word, the result of pronouncing each time is all different, also can not have identical time span.Therefore with when memory model matches, the time shaft of unknown words will distortion or bending unevenly, aligns to make its feature and template characteristic.DTW is very similar for waveform shape on the whole, but is the waveform do not lined up on a timeline, has good robustness.DTW, by time series is carried out extending and shortening, calculates the similarity between two time serieses.DTW is the corresponding time relationship describing test template and reference template with Time alignment function W (n) meeting certain condition, cumulative distance minimum corresponding warping function when solving two template matches.
Concrete, finds a best Time alignment function, make described row projection value curve closest to standard row projection value curve, the minor increment summation of acquisition match point time.In the present embodiment, the characteristic time sequence arranging a reference template is A={a
1, a
2..., a
i, input security features time series B={b to be identified
1, b
2..., b
i; wherein i ≠ j; if Time alignment function is C={c (1); c (2) ..., c (N) }; wherein N is path; c (n)=(i (n), j (n)) represents that the n-th matching double points is the matching double points be made up of the individual eigenvector of jth (n) of i-th (n) of reference template individual eigenvector and template to be identified, distance (or distortion value) d (a therebetween
i (n), b
i (n)) being called local matching distance, the minor increment summation of match point time is
Step S104, according to matching result, differentiates that light becomes the true and false of hollow out safety line.
In the present embodiment, it is threshold range time genuine that preset light becomes hollow out safety line, is set to 0.5-0.8, if D is ∈ [0.5,0.8], then light becomes hollow out safety line into true; If D > 0.8 or D < 0.5, then light becomes hollow out safety line into false.Those skilled in the art should know, light can be caused to become the corresponding change into threshold range time genuine produces of hollow out safety line according to the difference of image scanning instrument scanning accuracy, thus adopt different scanning accuracies to have different threshold ranges, do not repeat them here.
In sum, in the present embodiment, the recognition methods of bank note comprises carries out false distinguishing to the light change hollow out safety line of bank note.Become hollow out safety line position by the light obtaining bank note, hollow out safety line position is become to described light and carries out binaryzation; Calculate the row projection value that the light after binaryzation becomes hollow out safety line position, obtain row projection value curve according to described row projection value; Use row projection value curve described in dynamic time warping algorithmic match; According to matching result, differentiate that light becomes the true and false of hollow out safety line.The present invention carries out binaryzation to light change hollow out safety line position and obtains the row projection value of light change hollow out safety line position and obtain row projection value curve, dynamic time warping algorithmic match light is used to become hollow out safety line characteristic information, effectively can become hollow out safety line to the light of bank note and carry out false distinguishing identification, be conducive to the true and false prevention that light becomes hollow out safety line.
Please refer to Fig. 2, it is the method flow diagram of the second embodiment of the recognition methods of a kind of bank note provided in the specific embodiment of the invention.As shown in the figure, this recognition methods, comprising:
Step S201, identify the Information of bank note, described Information is divided into: front forward, front oppositely, reverse side forward and reverse side reverse.
Concrete, can according to the crown word number magnetic image of bank note to bank note towards identifying, if crown word number magnetic image is in the lower left corner of bank note, then bank note is front forward, if crown word number magnetic image is in the upper right corner of bank note, then bank note is that front is reverse, if crown word number magnetic image is at the lower right corner inferior horn of bank note, then bank note is reverse side forward, if crown word number magnetic image is in the upper left corner of bank note, then bank note is that reverse side is reverse.
Step S202, obtains the infrared transmission image of bank note, chooses light become hollow out safety line position according to described infrared transmission image, cuts the region subgraph that described light becomes hollow out safety line position.
Concrete, when the Information identifying bank note is front forward or reverse side is reverse, light becomes hollow out safety line position on the right side of bank note, and near the position of border, the right 3-5cm, namely the region subgraph cutting described light change hollow out safety line position cuts the image of the position of the right border 3-5cm; When the Information identifying bank note be front oppositely or reverse side forward time, light becomes hollow out safety line position in the left side of bank note, near the position of left boundary 3-5cm, namely the region subgraph cutting described light change hollow out safety line position cuts the image of the position of left boundary 3-5cm.
Step S203, calculates the row projection value of described region subgraph, extracts according to described row projection value the row projection value feature that light becomes hollow out safety line position.
In the present embodiment, Fig. 3 is the bank note schematic diagram of the miscellaneous editions provided in the specific embodiment of the invention, wherein a is 1999 millboard coin, b is 2005 millboard coin, c is 2015 millboard coin, Fig. 4 is the region subgraph row projection value curve map of the light change hollow out safety line position of the bank note of miscellaneous editions in Fig. 3, wherein a is 1999 editions row projection value curves, b is 2005 editions row projection value curves, c is 2015 editions row projection value curves, view field is the projection value of the region subgraph of cutting herein, namely light becomes the projection value in hollow out safety line region, the horizontal ordinate of row projection value represents the width coordinate of bank note, ordinate represent projection value and.There is light and become hollow out safety line in 2015 editions 100 yuan notes, the row projection value curve of its row projection value image and 1999 editions, 2005 editions is widely different.2015 editions 100 yuan notes become hollow out safety line owing to there is light, and light becomes the word printed words of existence 100 in hollow out safety line, and thus row projection value curve has the curve shape feature of an approximate U.
Step S204, identifying bank note according to the row projection value feature of described light change hollow out safety line position is have the bank note that light becomes hollow out safety line version.
Concrete, have the difference that bank note that light becomes hollow out safety line version and not light becomes the row projection value curve of the bank note of hollow out safety line version very large, whether by extracting corresponding row projection value curvilinear characteristic, can to identify bank note be easily the bank note with light change hollow out safety line version.In the present embodiment, 100 yuan note row projection value curves of various version as shown in Figure 4, only have 2015 editions 100 yuan notes in c to there is light and become hollow out safety line, the curve shape feature of an approximate U is had in its row projection value curve, can judge that light becomes hollow out safety line and whether exists according to whether there is this feature in row projection value curve, and accurately locate light according to the light 100 yuan of printed words become in hollow out safety line and become hollow out safety line position, thus pick out bank note and whether belong to 2015 editions 100 yuan notes.In addition, bank note is when crossing paper money, and same column data is through the same position of cmos image sensor (CIS), so row projected image gray scale relative different is less.Version identification because for be row projection properties, so Algorithm robustness is stronger.Those skilled in the art it should be known that light becomes the different or different meeting in position of word in hollow out safety line and makes safe alignment projection value curve present different characteristic, in like manner can judge that light becomes hollow out safety line and whether exists according to corresponding feature.
Step S205, accurately locates light according to described row projection value feature and becomes hollow out safety line position.
Concrete, there is the difference that bank note that light becomes hollow out safety line version and not light becomes the row projection value curve of the bank note of hollow out safety line version very large, can accurately locating light change hollow out safety line position by extracting corresponding row projection value curvilinear characteristic.In the present embodiment, as shown in c in Fig. 4,2015 editions 100 yuan notes become hollow out safety line owing to there is light, and light becomes the word printed words of existence 100 in hollow out safety line, thus row projection value curve has the curve shape feature of an approximate U, accurately can locate light become hollow out safety line position according to this curvilinear characteristic.Those skilled in the art it should be known that light becomes the different or different meeting in position of word in hollow out safety line and makes safe alignment projection value curve present different characteristic, in like manner accurately judge that light becomes the position of hollow out safety line according to corresponding feature.
Step S206, uses Adaptive Thresholding to become hollow out safety line position to described light and carries out binaryzation.
Concrete, Adaptive Thresholding is that whole region light being become hollow out safety line position is divided into three parts by word: the left side, centre, the right, middle part is the region of word, try to achieve the mean value of the left side and the right gray scale as gray threshold, being black when gray scale is more than or equal to gray threshold, is white when gray scale is less than gray threshold, thus completes the binaryzation that light becomes hollow out safety line position, effectively can carry out denoising to background, greatly simplify the complexity of calculating.In the present embodiment, hollow out safety line position is become to the light of 2015 editions 100 yuan notes and carries out the image after binaryzation as shown in Figure 5.
Step S207, calculates the row projection value that the light after binaryzation becomes hollow out safety line position, obtains row projection value curve according to described row projection value.
Concrete, bank note infrared view width is 200DPI resolution, be highly 150DPI resolution, the gray-scale value sum that light becomes every a line in hollow out safety line forms a row projection value, the row projection value that whole light becomes hollow out safety line position forms row projection value curve, after binaryzation, projection value can be comparatively large when running into font for safety line row projection value curve, and when not having font, row projection value can be less, row projection value curve be with font with or without and font change and change.In the present embodiment, the image after carrying out binaryzation to the light change hollow out safety line position of 2015 editions 100 yuan notes calculates its row projection value, and the projection value curve obtained as shown in Figure 6.
Step S208, uses row projection value curve described in dynamic time warping algorithmic match.
Step S209, according to matching result, differentiates that light becomes the true and false of hollow out safety line.
In sum, in the present embodiment, the recognition methods of bank note also comprises the bank note identifying and have light change hollow out safety line version.The present embodiment utilizes magnetic stripe transmittancy, in infrared transmission image, it is more weak that light becomes hollow out safety line transmittancy, and light becomes the feature that in hollow out safety line, word transmittancy is stronger, the row projection value curvilinear characteristic becoming hollow out safety line region according to light can identify bank note accurately whether into there is the bank note version of light change hollow out safety line; The present embodiment is use safety line provincial characteristics identification bank note version only, decreases algorithm calculated amount, improves algorithm recognition speed; Use dynamic time warping algorithm to extract safety line characteristic information, false distinguishing identification is carried out to new edition bank note safety line, be conducive to the prevention of the safety line true and false.
It is below the embodiment of the recognition device of a kind of bank note of this programme, a kind of embodiment of recognition device of bank note realizes based on a kind of embodiment of recognition methods of bank note, description not most in a kind of embodiment of recognition device of bank note, please refer to a kind of embodiment of recognition methods of bank note.
Please refer to Fig. 7, it is the block diagram of the first embodiment of the recognition device of a kind of bank note provided in the specific embodiment of the invention.As shown in the figure, this recognition device, comprising:
Acquiring unit 710, becomes hollow out safety line position for the light obtaining bank note, becomes hollow out safety line position carry out binaryzation to described light.
Concrete, the word in the middle of hollow out safety line can be become by display light clearly by infrared transmission image, adopt Adaptive Thresholding to become hollow out safety line position to light according to word in the distribution that light becomes in hollow out safety line and carry out binaryzation.Wherein, Adaptive Thresholding is that whole region light being become hollow out safety line position is divided into three parts by word: the left side, centre, the right, middle part is the region of word, try to achieve the mean value of the left side and the right gray scale as gray threshold, being black when gray scale is more than or equal to gray threshold, is white when gray scale is less than gray threshold, thus completes the binaryzation that light becomes hollow out safety line position, effectively can carry out denoising to background, greatly simplify the complexity of calculating.
Row projecting cell 720, becomes the row projection value of hollow out safety line position for calculating the light after binaryzation, obtain row projection value curve according to described row projection value.
Wherein, bank note infrared view width is 200DPI resolution, be highly 150DPI resolution, the gray-scale value sum that light becomes every a line in hollow out safety line forms a row projection value, the row projection value that whole light becomes hollow out safety line position forms row projection value curve, after binaryzation, projection value can be comparatively large when running into font for safety line row projection value curve, and when not having font, projection value can be less, projection value curve be with font with or without and font change and change.
Matching unit 730, for using row projection value curve described in dynamic time warping algorithmic match.
Wherein, dynamic time warping algorithm (DynamicTimeWarping, DTW) thought is: because voice signal is a kind of signal with quite large randomness, even if identical speaker is to identical word, the result of pronouncing each time is all different, also can not have identical time span.Therefore with when memory model matches, the time shaft of unknown words will distortion or bending unevenly, aligns to make its feature and template characteristic.DTW is very similar for waveform shape on the whole, but is the waveform do not lined up on a timeline, has good robustness.DTW, by time series is carried out extending and shortening, calculates the similarity between two time serieses.DTW is the corresponding time relationship describing test template and reference template with Time alignment function W (n) meeting certain condition, cumulative distance minimum corresponding warping function when solving two template matches.
Concrete, finds a best Time alignment function, make described row projection value curve closest to standard row projection value curve, the minor increment summation of acquisition match point time.In the present embodiment, the characteristic time sequence arranging a reference template is A={a
1, a
2..., a
i, input security features time series B={b to be identified
1, b
2..., b
j; wherein i ≠ j; if Time alignment function is C={c (1); c (2) ..., c (N) }; wherein N is path; c (n)=(i (n), j (n)) represents that the n-th matching double points is the matching double points be made up of the individual eigenvector of jth (n) of i-th (n) of reference template individual eigenvector and template to be identified, distance (or distortion value) d (a therebetween
i (n), b
i (n)) being called local matching distance, the minor increment summation of match point time is
False distinguishing unit 740, for according to matching result, differentiates that light becomes the true and false of hollow out safety line.
In the present embodiment, threshold range time preset word safety line is genuine, is set to 0.5-0.8, if D is ∈ [0.5,0.8], then word safety line is true; If D > 0.8 or D < 0.5, then word safety line is false.Those skilled in the art it should be known that light can be caused to become hollow out safety line according to the difference of image scanning instrument scanning accuracy produces change accordingly into threshold range time genuine, thus may occur different threshold ranges, not repeat herein.
In sum, above-mentioned each unit collaborative work, becomes hollow out safety line position by the light obtaining bank note, becomes hollow out safety line position carry out binaryzation to described light; Calculate the row projection value that the light after binaryzation becomes hollow out safety line position, obtain row projection value curve according to described row projection value; Use row projection value curve described in dynamic time warping algorithmic match; According to matching result, differentiate that light becomes the true and false of hollow out safety line.The present invention carries out binaryzation to light change hollow out safety line position and obtains the row projection value of light change hollow out safety line position and obtain row projection value curve, dynamic time warping algorithmic match light is used to become hollow out safety line characteristic information, effectively can become hollow out safety line to the light of bank note and carry out false distinguishing identification, be conducive to the true and false prevention that light becomes hollow out safety line.
Please refer to Fig. 8, it is the block diagram of the second embodiment of the recognition device of a kind of bank note provided in the specific embodiment of the invention.As shown in the figure, this recognition device, comprising:
Acquiring unit 710, becomes hollow out safety line position for the light obtaining bank note, becomes hollow out safety line position carry out binaryzation to described light.
Cutter unit 711, for obtaining the infrared transmission image of bank note, choosing light according to described infrared transmission image and becoming hollow out safety line position, cut the region subgraph that described light becomes hollow out safety line position.
Concrete, when the Information identifying bank note is front forward or reverse side is reverse, light becomes hollow out safety line position on the right side of bank note, and near the position of border, the right 3-5cm, namely the region subgraph cutting described light change hollow out safety line position cuts the image of the position of the right border 3-5cm; When the Information identifying bank note be front oppositely or reverse side forward time, light becomes hollow out safety line position in the left side of bank note, near the position of left boundary 3-5cm, namely the region subgraph cutting described light change hollow out safety line position cuts the image of the position of left boundary 3-5cm.
Row projecting cell 712, for calculating the row projection value of described region subgraph, extracts according to described row projection value the row projection value feature that light becomes hollow out safety line position.
In the present embodiment, Fig. 3 is the bank note schematic diagram of the miscellaneous editions provided in the specific embodiment of the invention, wherein a is 1999 millboard coin, b is 2005 millboard coin, c is 2015 millboard coin, Fig. 4 is the region subgraph row projection value curve map of the light change hollow out safety line position of the bank note of miscellaneous editions in Fig. 3, wherein a is 1999 editions row projection value curves, b is 2005 editions row projection value curves, c is 2015 editions row projection value curves, view field is the projection value of the region subgraph of cutting herein, namely light becomes the projection value in hollow out safety line region, the horizontal ordinate of row projection value represents the width coordinate of bank note, ordinate represent projection value row and.There is light and become hollow out safety line in obvious 2015 editions 100 yuan notes, the row projection value curve of its row projection value image and 1999 editions, 2005 editions is widely different.Wherein, 2015 editions 100 yuan notes become hollow out safety line owing to there is light, and light becomes the word printed words of existence 100 in hollow out safety line, and thus row projection value curve has an approximate U curve shape feature.
Positioning unit 713, becomes hollow out safety line position for accurately locating light according to described row projection value feature.
Specifically, according to whether there is row projection value feature, judge that light becomes hollow out safety line and whether exists, accurately location light becomes hollow out safety line position, thus distinguishes whether bank note belongs to the bank note that there is light change hollow out safety line version easily.In addition, bank note is when crossing paper money, and same column data is through the same position of cmos image sensor (CIS), so row projected image gray scale relative different is less.Version identification because for be row projection properties, so Algorithm robustness is stronger.
In the present embodiment, there is light and become hollow out safety line in 2015 editions 100 yuan notes, and light becomes the word printed words of existence 100 in hollow out safety line, its row projection value curve has the curve shape feature of an approximate U, accurately locates light become hollow out safety line position according to this curve shape feature.
Binarization unit 714, becomes hollow out safety line position for using Adaptive Thresholding to described light and carries out binaryzation.
Concrete, Adaptive Thresholding is that whole region light being become hollow out safety line position is divided into three parts by word: the left side, centre, the right, middle part is the region of word, try to achieve the mean value of the left side and the right gray scale as gray threshold, being black when gray scale is more than or equal to gray threshold, is white when gray scale is less than gray threshold, thus completes the binaryzation that light becomes hollow out safety line position, effectively can carry out denoising to background, greatly simplify the complexity of calculating.
Row projecting cell 720, becomes the row projection value of hollow out safety line position for calculating the light after binaryzation, obtain row projection value curve according to described row projection value.
Matching unit 730, for using row projection value curve described in dynamic time warping algorithmic match.
False distinguishing unit 740, for according to matching result, differentiates that light becomes the true and false of hollow out safety line.
Towards recognition unit 750, for identifying the Information of bank note, described Information is divided into: front forward, front oppositely, reverse side forward and reverse side reverse.
Concrete, can according to the crown word number magnetic image of bank note to bank note towards identifying, if crown word number magnetic image is in the lower left corner of bank note, then bank note is front forward, if crown word number magnetic image is in the upper right corner of bank note, then bank note is that front is reverse, if crown word number magnetic image is at the lower right corner inferior horn of bank note, then bank note is reverse side forward, if crown word number magnetic image is in the upper left corner of bank note, then bank note is that reverse side is reverse.
Version recognition unit 760, it is have the bank note that light becomes hollow out safety line version that the row projection value feature for becoming hollow out safety line position according to described light identifies bank note.
In sum, the present embodiment utilizes magnetic stripe transmittancy, and in infrared transmission image, safety line transmittancy is more weak, and the feature that in safety line, word transmittancy is stronger, whether identify bank note is accurately there is the bank note version that light becomes hollow out safety line; The present embodiment is use safety line provincial characteristics identification bank note version only, decreases algorithm calculated amount, improves algorithm recognition speed; The present embodiment uses dynamic time warping algorithm to extract safety line characteristic information, carries out false distinguishing identification to new edition bank note safety line, is conducive to the prevention of the safety line true and false.
Training in relative neural network BP training algorithm and identification, the algorithm in the embodiment of the present invention only has recognizer, and algorithmic code is more easily safeguarded; Only use safety line provincial characteristics identification bank note version, decreases algorithm calculated amount, improves algorithm recognition speed; Do not rely on the brightness that CIS sensor gathers image, weaken variation of image grayscale to the impact of version identification, utilize magnetic stripe transmittancy, in infrared transmission image, safety line transmittancy is more weak, and the feature that in safety line, word transmittancy is stronger, whether cog region divides bank note to be that light becomes hollow out safety line version bank note, improves version discrimination; Use dynamic time warping algorithmic match safety line characteristic information to carry out false distinguishing identification to 2015 editions bank note safety lines in addition, be conducive to the prevention of the safety line true and false.
Below know-why of the present invention is described in conjunction with specific embodiments.These describe just in order to explain principle of the present invention, and can not be interpreted as limiting the scope of the invention by any way.Based on explanation herein, those skilled in the art does not need to pay performing creative labour can associate other embodiment of the present invention, and these modes all will fall within protection scope of the present invention.
Claims (10)
1. a recognition methods for bank note, is characterized in that, comprising:
The light obtaining bank note becomes hollow out safety line position, becomes hollow out safety line position carry out binaryzation to described light;
Calculate the row projection value that the light after binaryzation becomes hollow out safety line position, obtain row projection value curve according to described row projection value;
Use row projection value curve described in dynamic time warping algorithmic match;
According to matching result, differentiate that light becomes the true and false of hollow out safety line.
2. method according to claim 1, is characterized in that, described acquisition light becomes hollow out safety line position, becomes hollow out safety line position and carries out binaryzation, comprising described light:
Obtain the infrared transmission image of bank note, choose light according to described infrared transmission image and become hollow out safety line position, cut the region subgraph that described light becomes hollow out safety line position;
Calculate the row projection value of described region subgraph, extract according to described row projection value the row projection value feature that light becomes hollow out safety line position;
Accurately locate light according to described row projection value feature and become hollow out safety line position;
Use Adaptive Thresholding to become hollow out safety line position to described light and carry out binaryzation.
3. method according to claim 2, it is characterized in that, the row projection value of described calculating described region subgraph, extract the projection value feature of light change hollow out safety line position according to described row projection value after, describedly accurately locate light according to described projection value feature and become before hollow out safety line position, also comprise:
Identifying bank note according to the row projection value feature of described light change hollow out safety line position is have the bank note that light becomes hollow out safety line version.
4. method according to claim 1, is characterized in that, the light of described acquisition bank note becomes hollow out safety line position, becomes before hollow out safety line position carries out binaryzation, also comprise described light:
Identify the Information of bank note, described Information is divided into: front forward, front oppositely, reverse side forward and reverse side reverse;
The light of described acquisition bank note becomes hollow out safety line position, becomes hollow out safety line position and carries out binaryzation, be specially described light:
When the Information identifying bank note is front forward or reverse side is reverse, the light obtaining bank note from the right side of bank note becomes hollow out safety line position, becomes hollow out safety line position carry out binaryzation to described light;
When the Information identifying bank note be front oppositely or reverse side forward time, the light obtaining bank note from the left side of bank note becomes hollow out safety line position, becomes hollow out safety line position carry out binaryzation to described light.
5. method according to claim 1, it is characterized in that, row projection value curve described in described use dynamic time warping algorithmic match, is specially:
Find a best Time alignment function, make described row projection value curve closest to standard row projection value curve, obtain the minor increment summation of match point time.
6. method according to claim 5, is characterized in that, described according to matching result, differentiates that light becomes the true and false of hollow out safety line, is specially:
If the minor increment summation of described match point time is within the scope of preset distance, then described light becomes hollow out safety line into true, otherwise is false.
7. a recognition device for bank note, is characterized in that, comprising:
Acquiring unit, becomes hollow out safety line position for the light obtaining bank note, becomes hollow out safety line position carry out binaryzation to described light;
Row projecting cell, becomes the row projection value of hollow out safety line position for calculating the light after binaryzation, obtain row projection value curve according to described row projection value;
Matching unit, for using row projection value curve described in dynamic time warping algorithmic match;
False distinguishing unit, for according to matching result, differentiates that light becomes the true and false of hollow out safety line.
8. device according to claim 7, is characterized in that, also comprises:
Towards recognition unit, for identifying the Information of bank note, described Information is divided into: front forward, front oppositely, reverse side forward and reverse side reverse.
9. device according to claim 7, is characterized in that, also comprises:
Version recognition unit, it is have the bank note that light becomes hollow out safety line version that the row projection value feature for becoming hollow out safety line position according to described light identifies bank note.
10. device according to claim 7, is characterized in that, described acquiring unit comprises:
Cutter unit, for obtaining the infrared transmission image of bank note, choosing light according to described infrared transmission image and becoming hollow out safety line position, cut the region subgraph that described light becomes hollow out safety line position;
Row projecting cell, for calculating the row projection value of described region subgraph, extracts according to described row projection value the row projection value feature that light becomes hollow out safety line position;
Positioning unit, becomes hollow out safety line position for accurately locating light according to described row projection value feature;
Binarization unit, becomes hollow out safety line position for using Adaptive Thresholding to described light and carries out binaryzation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510506975.6A CN105184955B (en) | 2015-08-18 | 2015-08-18 | A kind of recognition methods of bank note and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510506975.6A CN105184955B (en) | 2015-08-18 | 2015-08-18 | A kind of recognition methods of bank note and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105184955A true CN105184955A (en) | 2015-12-23 |
| CN105184955B CN105184955B (en) | 2018-03-02 |
Family
ID=54907003
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510506975.6A Expired - Fee Related CN105184955B (en) | 2015-08-18 | 2015-08-18 | A kind of recognition methods of bank note and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105184955B (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719386A (en) * | 2016-01-18 | 2016-06-29 | 新达通科技股份有限公司 | Paper currency identification method and device based on optically-variable hollowed windowing security line |
| CN106251467A (en) * | 2016-08-05 | 2016-12-21 | 深圳怡化电脑股份有限公司 | The detection method of a kind of safety line magnetic feature and device |
| CN106373256A (en) * | 2016-08-23 | 2017-02-01 | 深圳怡化电脑股份有限公司 | Method and system for RMB version recognition |
| CN106408746A (en) * | 2016-08-25 | 2017-02-15 | 深圳怡化电脑股份有限公司 | Safety thread identification method and apparatus |
| CN106447910A (en) * | 2016-09-08 | 2017-02-22 | 深圳怡化电脑股份有限公司 | Paper money identifying method and apparatus |
| CN106504406A (en) * | 2016-11-01 | 2017-03-15 | 深圳怡化电脑股份有限公司 | A kind of method and device of identification bank note |
| CN106652166A (en) * | 2016-11-01 | 2017-05-10 | 深圳怡化电脑股份有限公司 | Detection method and system for overlapped banknotes |
| CN107195070A (en) * | 2017-05-18 | 2017-09-22 | 深圳怡化电脑股份有限公司 | A kind of secret note authentication method and device |
| CN107274543A (en) * | 2017-06-23 | 2017-10-20 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note, device, terminal device and computer-readable storage medium |
| CN107369243A (en) * | 2017-06-23 | 2017-11-21 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note, device, terminal device and computer-readable storage medium |
| CN107610320A (en) * | 2017-09-06 | 2018-01-19 | 深圳怡化电脑股份有限公司 | A kind of bank slip recognition method and apparatus |
| CN107610318A (en) * | 2017-08-23 | 2018-01-19 | 宁波术有电子科技有限公司 | 2015 editions RMB, which are carried out, using the imaging of metal wire backlight discriminates method for distinguishing |
| CN108022363A (en) * | 2016-11-02 | 2018-05-11 | 深圳怡化电脑股份有限公司 | A kind of bank note towards recognition methods and device |
| CN108074323A (en) * | 2016-11-15 | 2018-05-25 | 深圳怡化电脑股份有限公司 | A kind of bank note towards recognition methods and its device |
| CN108074322A (en) * | 2016-11-17 | 2018-05-25 | 深圳怡化电脑股份有限公司 | A kind of image-recognizing method and device |
| CN109300220A (en) * | 2018-08-29 | 2019-02-01 | 广州广电运通金融电子股份有限公司 | Paper false distinguishing method, apparatus, equipment and readable storage medium storing program for executing based on safety line |
| CN109697443A (en) * | 2017-10-23 | 2019-04-30 | 深圳怡化电脑股份有限公司 | A kind of paper money number dividing method and splitting equipment |
| CN110956737A (en) * | 2020-01-07 | 2020-04-03 | 武汉卓目科技有限公司 | Safety line identification method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009113406A1 (en) * | 2008-03-14 | 2009-09-17 | アルゼ株式会社 | Paper identifying apparatus and paper identifying method |
| CN101908241A (en) * | 2010-08-03 | 2010-12-08 | 广州广电运通金融电子股份有限公司 | Method and system for identifying valued documents |
| CN102096962A (en) * | 2010-12-23 | 2011-06-15 | 北京新岸线软件科技有限公司 | Paper currency detecting method and device |
| CN102737435A (en) * | 2011-04-11 | 2012-10-17 | 北京新岸线数字图像技术有限公司 | Paper money discrimination method and device |
| CN104091388A (en) * | 2014-07-22 | 2014-10-08 | 新达通科技股份有限公司 | Paper currency authentic identification method and device based on magnetic images |
| CN104298989A (en) * | 2014-08-20 | 2015-01-21 | 深圳大学 | Counterfeit identifying method and counterfeit identifying system based on zebra crossing infrared image characteristics |
-
2015
- 2015-08-18 CN CN201510506975.6A patent/CN105184955B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009113406A1 (en) * | 2008-03-14 | 2009-09-17 | アルゼ株式会社 | Paper identifying apparatus and paper identifying method |
| CN101908241A (en) * | 2010-08-03 | 2010-12-08 | 广州广电运通金融电子股份有限公司 | Method and system for identifying valued documents |
| CN102096962A (en) * | 2010-12-23 | 2011-06-15 | 北京新岸线软件科技有限公司 | Paper currency detecting method and device |
| CN102737435A (en) * | 2011-04-11 | 2012-10-17 | 北京新岸线数字图像技术有限公司 | Paper money discrimination method and device |
| CN104091388A (en) * | 2014-07-22 | 2014-10-08 | 新达通科技股份有限公司 | Paper currency authentic identification method and device based on magnetic images |
| CN104298989A (en) * | 2014-08-20 | 2015-01-21 | 深圳大学 | Counterfeit identifying method and counterfeit identifying system based on zebra crossing infrared image characteristics |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719386A (en) * | 2016-01-18 | 2016-06-29 | 新达通科技股份有限公司 | Paper currency identification method and device based on optically-variable hollowed windowing security line |
| CN106251467A (en) * | 2016-08-05 | 2016-12-21 | 深圳怡化电脑股份有限公司 | The detection method of a kind of safety line magnetic feature and device |
| CN106251467B (en) * | 2016-08-05 | 2019-03-12 | 深圳怡化电脑股份有限公司 | A kind of detection method and device of safety line magnetic feature |
| CN106373256B (en) * | 2016-08-23 | 2019-04-26 | 深圳怡化电脑股份有限公司 | The method and system of RMB version identification |
| CN106373256A (en) * | 2016-08-23 | 2017-02-01 | 深圳怡化电脑股份有限公司 | Method and system for RMB version recognition |
| CN106408746A (en) * | 2016-08-25 | 2017-02-15 | 深圳怡化电脑股份有限公司 | Safety thread identification method and apparatus |
| CN106408746B (en) * | 2016-08-25 | 2019-03-12 | 深圳怡化电脑股份有限公司 | A kind of safety line recognition methods and device |
| CN106447910A (en) * | 2016-09-08 | 2017-02-22 | 深圳怡化电脑股份有限公司 | Paper money identifying method and apparatus |
| CN106447910B (en) * | 2016-09-08 | 2019-07-05 | 深圳怡化电脑股份有限公司 | A kind of method and device of paper money recognition |
| CN106652166A (en) * | 2016-11-01 | 2017-05-10 | 深圳怡化电脑股份有限公司 | Detection method and system for overlapped banknotes |
| CN106652166B (en) * | 2016-11-01 | 2019-06-07 | 深圳怡化电脑股份有限公司 | A kind of detection method and system of stuck banknotes |
| CN106504406B (en) * | 2016-11-01 | 2019-11-15 | 深圳怡化电脑股份有限公司 | A kind of method and device identifying bank note |
| CN106504406A (en) * | 2016-11-01 | 2017-03-15 | 深圳怡化电脑股份有限公司 | A kind of method and device of identification bank note |
| CN108022363A (en) * | 2016-11-02 | 2018-05-11 | 深圳怡化电脑股份有限公司 | A kind of bank note towards recognition methods and device |
| CN108074323A (en) * | 2016-11-15 | 2018-05-25 | 深圳怡化电脑股份有限公司 | A kind of bank note towards recognition methods and its device |
| CN108074322A (en) * | 2016-11-17 | 2018-05-25 | 深圳怡化电脑股份有限公司 | A kind of image-recognizing method and device |
| CN107195070B (en) * | 2017-05-18 | 2019-09-20 | 深圳怡化电脑股份有限公司 | A kind of secret note identification method and device |
| CN107195070A (en) * | 2017-05-18 | 2017-09-22 | 深圳怡化电脑股份有限公司 | A kind of secret note authentication method and device |
| CN107369243A (en) * | 2017-06-23 | 2017-11-21 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note, device, terminal device and computer-readable storage medium |
| CN107274543A (en) * | 2017-06-23 | 2017-10-20 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note, device, terminal device and computer-readable storage medium |
| CN107369243B (en) * | 2017-06-23 | 2019-08-27 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note, device, terminal device and computer storage medium |
| CN107610318A (en) * | 2017-08-23 | 2018-01-19 | 宁波术有电子科技有限公司 | 2015 editions RMB, which are carried out, using the imaging of metal wire backlight discriminates method for distinguishing |
| CN107610318B (en) * | 2017-08-23 | 2019-12-06 | 成都术有科技有限公司 | Method for discriminating 2015 edition RMB by utilizing metal wire backlight imaging |
| CN107610320A (en) * | 2017-09-06 | 2018-01-19 | 深圳怡化电脑股份有限公司 | A kind of bank slip recognition method and apparatus |
| CN109697443A (en) * | 2017-10-23 | 2019-04-30 | 深圳怡化电脑股份有限公司 | A kind of paper money number dividing method and splitting equipment |
| CN109300220A (en) * | 2018-08-29 | 2019-02-01 | 广州广电运通金融电子股份有限公司 | Paper false distinguishing method, apparatus, equipment and readable storage medium storing program for executing based on safety line |
| CN110956737A (en) * | 2020-01-07 | 2020-04-03 | 武汉卓目科技有限公司 | Safety line identification method and device |
| CN110956737B (en) * | 2020-01-07 | 2021-10-12 | 武汉卓目科技有限公司 | Safety line identification method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105184955B (en) | 2018-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105184955A (en) | Method and device for identifying paper money | |
| CN108090429B (en) | Vehicle type recognition method for graded front face bayonet | |
| CN101350063B (en) | Method and apparatus for locating human face characteristic point | |
| CN104298982B (en) | A kind of character recognition method and device | |
| CN102324132B (en) | Bank-note detecting method by identifying ultraviolet-light image and system | |
| CN102542655B (en) | Note anti-counterfeiting discrimination method based on fiber personality characteristics | |
| CN104318238A (en) | Method for extracting crown word numbers from scanned banknote images in banknote detection module | |
| CN103778409A (en) | Human face identification method based on human face characteristic data mining and device | |
| CN106557764A (en) | A kind of water level recognition methodss based on binary-coded character water gauge and image procossing | |
| CN104464079A (en) | Multi-currency-type and face value recognition method based on template feature points and topological structures of template feature points | |
| CN104091388A (en) | Paper currency authentic identification method and device based on magnetic images | |
| CN104680130A (en) | Chinese character recognition method for identification cards | |
| CN108235115B (en) | Method and terminal for positioning human voice area in song video | |
| CN105761351A (en) | Structure characteristic-based character recognition method | |
| CN102156884A (en) | Straight segment detecting and extracting method | |
| CN104077594A (en) | Image recognition method and device | |
| CN105160756A (en) | Paper money facing direction recognition method and device | |
| CN105243729A (en) | Method and system for identifying banknote version | |
| CN105426890A (en) | Method for identifying graphic verification code with twisty and adhesion characters | |
| CN104392229A (en) | Handwriting identification method based on stroke fragment direction distribution characteristics | |
| CN105574981A (en) | RMB banknote denomination identification method based on safety line magnetic information | |
| CN104537364A (en) | Dollar bill denomination and edition identifying method based on texture analysis | |
| CN106340116B (en) | A kind of recognition methods of bank note and device | |
| CN107742357A (en) | A kind of recognition methods of paper money number and device | |
| CN105139508A (en) | Method and device for detecting paper money |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180302 Termination date: 20210818 |