CN112735021B - Banknote zebra crossing identification and counterfeit detection method - Google Patents
Banknote zebra crossing identification and counterfeit detection method Download PDFInfo
- Publication number
- CN112735021B CN112735021B CN202011635134.2A CN202011635134A CN112735021B CN 112735021 B CN112735021 B CN 112735021B CN 202011635134 A CN202011635134 A CN 202011635134A CN 112735021 B CN112735021 B CN 112735021B
- Authority
- CN
- China
- Prior art keywords
- zebra crossing
- data
- zebra
- black
- stripes
- 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
- 241000283070 Equus zebra Species 0.000 title claims abstract description 91
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 230000005540 biological transmission Effects 0.000 claims abstract description 21
- 230000037303 wrinkles Effects 0.000 claims abstract description 7
- 238000000034 method Methods 0.000 claims description 13
- 238000012216 screening Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/06—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using wave or particle radiation
- G07D7/12—Visible light, infrared or ultraviolet radiation
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/20—Testing patterns thereon
- G07D7/2008—Testing patterns thereon using pre-processing, e.g. de-blurring, averaging, normalisation or rotation
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/20—Testing patterns thereon
- G07D7/2016—Testing patterns thereon using feature extraction, e.g. segmentation, edge detection or Hough-transformation
Abstract
The invention discloses a bank note zebra crossing identification and false detection method, which comprises the following steps: 1) acquiring an infrared transmission light image of the bank note, and extracting a gray level image of the infrared transmission light of an area where the zebra crossing is located; 2) calculating the average value of each row of pixels of the gray level image to form a group of data; 3) selecting a convolution kernel, and performing convolution operation on the group of data twice to obtain central point position data of the zebra crossing black and white bars; 4) processing the data of the central point of the zebra crossing black and white stripes, and removing the data which do not meet the requirements; 5) calculating the characteristic data of the zebra stripes according to the central point positions of the black and white stripes of the zebra stripes; 6) and comparing the zebra crossing characteristic data with preset data so as to judge whether the zebra crossing is true or false. The banknote zebra crossing identification and false detection method adopts convolution operation, and can quickly and effectively remove image noise generated by stains and wrinkles of the banknote and inconsistent image brightness generated by uneven infrared light source.
Description
Technical Field
The invention relates to a banknote identification and counterfeit detection technology, in particular to a banknote zebra crossing identification and counterfeit detection method.
Background
The currency counter is an electromechanical integrated device for automatically counting the number of the currency notes, and generally has a counterfeit currency identification function and integrates counting and counterfeit currency identification.
Because the domestic cash circulation scale is huge, cash processing work of enterprises and public institutions and financial institutions is heavy, and the cash counter becomes an indispensable device. With the development of printing technology, copying technology and electronic scanning technology, the manufacturing level of counterfeit money is higher and higher, and the counterfeit identification performance of the money counter must be continuously improved.
Since the promulgation of the gold mark of the RMB currency counting machine, a new round of development is brought to the currency counting machine, and the original detection technology needs to be improved urgently.
During the use and circulation of the bank note, the surface of the bank note is difficult to avoid the phenomena of stains, folds and the like, the brightness and the uniformity of infrared light of different bank note counting machines are different, and the factors have certain influence on infrared transmission images of the bank note counting machines. For a system adopting image recognition to detect false, it is important how to quickly and effectively remove noise and obtain more real zebra crossing parameters.
The existing method for detecting the zebra crossing by the cash register generally adopts an average algorithm and a threshold algorithm or a gradient algorithm aiming at the pixel value of an image, and the algorithms are difficult to avoid the influence of noise and stains and cannot accurately identify the width and the demarcation point of the black and white stripes of the zebra crossing.
Disclosure of Invention
Aiming at the defects of the prior cash register, the invention aims to provide a more effective, accurate and rapid method for identifying and detecting the false of the zebra crossing of the cash
In order to solve the technical problems, the invention adopts the technical scheme that:
the invention provides a bank note zebra crossing identification and false detection method, which comprises the following steps:
1) acquiring an infrared transmission light image of the bank note, and extracting a gray level image of the infrared transmission light of an area where the zebra crossing is located;
2) calculating the average value of each row of pixels of the gray level image to form a group of data;
3) selecting a convolution kernel, and performing convolution operation on the group of data twice to obtain central point position data of the zebra crossing black and white bars;
4) processing the data of the central point of the zebra crossing black and white stripes, and removing the data which do not meet the requirements;
5) calculating the characteristic data of the zebra stripes according to the central point positions of the black and white stripes of the zebra stripes;
6) and comparing the zebra crossing characteristic data with preset data so as to judge whether the zebra crossing is true or false.
The gray image in the step 1) is a zebra crossing gray image obtained by fixing the width and the height of the whole infrared transmission image of the bank note.
The group of data in the step 2) is an arithmetic average value obtained by dividing the sum of pixels of each row of the zebra crossing gray-scale image by the width, and the number of the arithmetic average values is an image height value.
Selecting a convolution kernel in the step 3) as follows: setting the height of the black stripes as x and the height of the white stripes as y, and setting the height of the zebra crossing as H to be x + y; by h H (b) 1, (b) 0,1,.. H-1) as the convolution kernel of the first convolution operation, and H H/2 (b) H/2-1) as the convolution kernel for the second convolution operation.
The convolution operation twice in the step 3) is as follows:
let the image height be W, and the convolution object of the first convolution operation be the arithmetic mean value f (t) of the pixels on the zebra crossing, where t is (0, 1.... W), that is, it is WWherein a is 0, 1.... W;
then carrying out difference operation f 1 (a) F (t) -g (a), eliminating the effect of luminance non-uniformity;
the convolution object of the second convolution operation is f 1 (a) I.e. by
The central point position of the black and white stripe of the zebra crossing obtained in the step 3) is set as f 2 (a) And (4) screening extreme values of data formed by the second convolution operation, wherein the maximum value is the central point of the white stripe of the zebra stripes, and the minimum value is the central point of the black stripe of the zebra stripes.
Processing the data of the central points of the zebra crossing black and white stripes in the step 4), namely, selecting the distance interval between two adjacent extreme points as [ H/2-delta, H/2+ delta ], and obtaining new data by screening [ H x 10% and H x 20% ].
The characteristic data of the zebra crossing in the step 5) comprises the following steps: black bar height dataset, white bar height dataset, number of black bar height approximations, number of white bar height approximations, and number of black and white alternate occurrences.
The invention has the following beneficial effects and advantages:
1. the banknote zebra crossing identification and false detection method adopts convolution operation, and can quickly and effectively remove image noise generated by stains and wrinkles of the banknote and inconsistent image brightness generated by uneven infrared light source.
2. The operation object of the invention is only the line pixel mean value of the image, the consumed time and resources are very limited, the invention is very beneficial to the whole currency counting machine false detection system, and the method can also be applied to other paper currency bills with similar characteristics.
Drawings
Fig. 1 is a control flow chart of the banknote zebra crossing identification and counterfeit detection method.
FIG. 2 is a gray scale representation of infrared transmission of a banknote in accordance with the method of the present invention;
FIG. 3 is a gray scale plot of zebra crossing extracted by the method of the present invention;
FIG. 4 is a graph of infrared transmission gray scale of a 100 RMB banknote issued in 2005;
FIG. 5 is a comparison graph of the convolution operation of the method for the genuine banknote of FIG. 4;
FIG. 6 is a graph of infrared transmission gray scale of a 100 RMB banknote issued 2005 by counterfeit banknotes (without zebra stripes);
FIG. 7 is a comparison graph of convolution operations performed on the counterfeit banknote of FIG. 6 according to the method and system of the present invention;
FIG. 8 is a graph of black and white stripe effects plotted against zebra stripes parameters obtained from a 2005 release 100-dollar banknote by the method of the present invention;
fig. 9 is a graph of black and white stripe effects plotted against zebra stripe parameters obtained by the method of the present invention for 100-yuan banknotes issued in 2015.
Detailed Description
The invention is further elucidated with reference to the accompanying drawings.
As shown in fig. 1, the invention provides a banknote zebra crossing identification and counterfeit detection method, which comprises the following steps:
1) acquiring an infrared transmission light image of the bank note, and extracting a gray level image of the infrared transmission light of an area where the zebra crossing is located;
2) calculating the average value of each row of pixels of the gray level image to form a group of data;
3) selecting a convolution kernel, and performing convolution operation on the group of data twice to obtain central point position data of the zebra crossing black and white bars;
4) processing the data of the central point of the zebra crossing black and white stripes, and removing the data which do not meet the requirements;
5) calculating the characteristic data of the zebra stripes according to the central point positions of the black and white stripes of the zebra stripes;
6) and comparing the zebra crossing characteristic data with preset data so as to judge whether the zebra crossing is true or false.
In the step 1), the infrared transmission gray image of the bank note is obtained, and is a zebra crossing gray image obtained by fixing the width and the height of the whole infrared transmission image of the bank note. And presetting zebra crossing parameters including black strip height data sets, white strip height data sets, the number of approximate black strip heights, the number of approximate white strip heights and the number of continuous black and white alternate occurrence times according to the image.
The group of data in the step 2) is an arithmetic average value obtained by dividing the sum of pixels of each row of the zebra crossing gray-scale image by the width, and the number of the arithmetic average values is an image height value.
Extracting an infrared transmission gray image of an area where the zebra crossing is located, and obtaining a two-dimensional array calibrated by using the height (row) W and the width (column) L of the image; calculating an arithmetic mean of all pixels of each row of the two-dimensional array, and obtaining a one-dimensional array f (t) calibrated to the image height (row) W, wherein t ═ 0, 1.... W);
selecting a convolution kernel in the step 3) as follows: setting the height of the black stripes as x and the height of the white stripes as y, and setting the height of the zebra crossing as H to be x + y; by h H (b) H-1 as the convolution kernel for the first convolution operation, with H being equal to 1, (b being equal to 0,1 H/2 (b) H/2-1) as the convolution kernel for the second convolution operation.
Calculating an arithmetic mean of all pixels of each row of the two-dimensional array, and obtaining a one-dimensional array f (t) calibrated to the image height (row) W, wherein t ═ 0, 1.... W);
for the first convolution operation, selecting a convolution kernel h H (b) H-1, the convolution target is f (t), i.e. 1, (b) 0,1Wherein a is 0, 1.... W;
difference operation, i.e. f 1 (a) F (t) -g (a), eliminating the effect of luminance non-uniformity;
the second convolution operation selects a convolution kernel h H/2 (b) H/2-1, the convolution object is f 1 (a) I.e. byEliminating noise generated by stains and wrinkles;
to f is paired 2 (a) Screening extreme values of the formed data, wherein the maximum value is the central point of a zebra crossing white stripe, and the minimum value is the central point of a zebra crossing black stripe;
processing the position data of the central point of the black and white stripes of the zebra stripes in the step 4), namely the distance interval between two adjacent extreme points is [ H/2-delta, H/2+ delta]Delta is [ H10%, H20% ]]If the distance is not reasonable, rejecting the new data f obtained after screening 3 (a);
In step 5) with f 3 (a) The data of (2) are taken as the basis to obtain the zebra crossing parameters, which mainly comprise: black bar height dataset, white bar height dataset, number of black bar height approximations, number of white bar height approximations, black and white intersectionAnd alternate for consecutive times.
As shown in fig. 2, a red transmission gray scale image of the acquired banknote is obtained in which black and white stripes of zebra stripes are visible.
As shown in fig. 3, the zebra crossing infrared transmission gray image is acquired at a fixed width and height, where W is the image height, L is the image width, x is the black stripe height, y is the white stripe height, and H is the zebra crossing height.
As shown in fig. 4, a clear zebra crossing can be seen in the infrared transmission gray scale diagram of the banknote of the 2005 edition rmb of the present invention, which is 100 yuan.
Fig. 5 is a comparison diagram after convolution operation is performed on the zebra crossing region shown in fig. 4 according to the present invention. The icon A is a projection drawing of the average value of the row pixels of the zebra stripes, the Y axis is the average value of the row pixels, and the X axis is the row number; certain data fluctuation is reflected due to the noise such as the uniformity of lamplight, stains, wrinkles and the like. And the icon B is a projection drawing after the first convolution and difference operation, and noise caused by lamp uniformity is eliminated. And the icon C is a projection image after the second convolution operation, so that noise such as dirt wrinkles is eliminated. The areas included by the vertical lines a and b and the white bar areas of the zebra stripes, the peak point of the projection graph of the icon C is the central point position of the white bar areas, and the valley point is the central point position of the black bar areas.
As shown in fig. 6, it is a 100 yuan infrared transmission gray scale diagram of counterfeit money (without zebra stripes) 2005 RMB collected by the present invention; it can be seen that the zebra crossing has been missing.
Fig. 7 is a comparison diagram of the false banknote (no zebra crossing) zebra crossing region of fig. 6 after convolution operation according to the present invention. Icon a is a projection of the average of the row pixels of the zebra stripes, the Y-axis being the row pixel average and the X-axis being the row number. Icon B is the projection after the first convolution and difference operation. It can be seen that the icon a and icon B data are almost irregularly found. The icon C is a projection graph after the second convolution operation, the data span of the vertical coordinate is very small and almost within +/-1, and the data span is not periodic, and the indexes can be used as the interpretation basis of the zebra-free lines.
As shown in fig. 8(2005 edition 100 yuan) and fig. 9(2015 edition 100 yuan), which are comparison graphs of the zebra crossing parameter plot and the actual image obtained by the method of the present invention, the degree of engagement is highly uniform.
Claims (6)
1. A bank note zebra crossing identification and counterfeit detection method is characterized by comprising the following steps:
1) acquiring an infrared transmission light image of the bank note, and extracting a gray level image of the infrared transmission light of an area where the zebra crossing is located;
2) calculating the average value of each row of pixels of the gray level image to form a group of data;
3) selecting a convolution kernel, and performing convolution operation on the group of data twice to obtain central point position data of the zebra crossing black and white bars;
selecting a convolution kernel in the step 3) as follows: setting the height of the black stripes as x and the height of the white stripes as y, and setting the height of the zebra crossing as H = x + y; to be provided withAs a convolution kernel of a first convolution operation toAs the convolution kernel of the second convolution operation;
the convolution operation twice in the step 3) is as follows:
let the image height be W, and the convolution object of the first convolution operation be the arithmetic mean of the pixels of the zebra crossingWherein t = (0, 1.... W), that is, t = (0, 1.... W) }Wherein a =0,1, ·.
4) processing the data of the central point of the zebra crossing black and white stripes, and removing the data which do not meet the requirements;
5) calculating the characteristic data of the zebra stripes according to the central point positions of the black and white stripes of the zebra stripes;
6) and comparing the zebra crossing characteristic data with preset data so as to judge whether the zebra crossing is true or false.
2. The banknote zebra crossing identification and counterfeit detection method of claim 1, wherein: the gray image in the step 1) is a zebra crossing gray image obtained by fixing the width and the height of the whole infrared transmission image of the bank note.
3. The banknote zebra crossing identification and counterfeit detection method of claim 1, wherein: the group of data in the step 2) is an arithmetic average value obtained by dividing the sum of pixels of each row of the zebra crossing gray-scale image by the width, and the number of the arithmetic average values is an image height value.
4. The banknote zebra crossing identification and counterfeit detection method of claim 1, wherein: obtaining the central point position of the black and white stripes of the zebra crossing in the step 3), andand (4) screening extreme values of data formed by the second convolution operation, wherein the maximum value is the central point of the white stripe of the zebra stripes, and the minimum value is the central point of the black stripe of the zebra stripes.
5. The method of claim 1, wherein the method comprises the step of identifying and verifying the zebra crossingThe method comprises the following steps: processing the position data of the central point of the black and white stripes of the zebra crossing in the step 4), namely the distance interval between two adjacent extreme points is,GetAnd obtaining new data through screening.
6. The banknote zebra crossing identification and counterfeit detection method of claim 1, wherein: the characteristic data of the zebra crossing in the step 5) comprises the following steps: black bar height dataset, white bar height dataset, number of black bar height approximations, number of white bar height approximations, and number of consecutive black and white alternate occurrences.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011635134.2A CN112735021B (en) | 2020-12-31 | 2020-12-31 | Banknote zebra crossing identification and counterfeit detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011635134.2A CN112735021B (en) | 2020-12-31 | 2020-12-31 | Banknote zebra crossing identification and counterfeit detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112735021A CN112735021A (en) | 2021-04-30 |
CN112735021B true CN112735021B (en) | 2022-08-16 |
Family
ID=75608609
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011635134.2A Active CN112735021B (en) | 2020-12-31 | 2020-12-31 | Banknote zebra crossing identification and counterfeit detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112735021B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5093184A (en) * | 1989-06-02 | 1992-03-03 | Portals Limited | Security paper with metallic patterned elongated security element |
DE102005016807A1 (en) * | 2005-04-05 | 2006-10-12 | Adolf Wielers | Security strip for incorporation in money paper, to combat counterfeiting of banknotes, is provided and/or printed with an electronically readable code or electrically conductive chemical |
CN102926280A (en) * | 2012-10-25 | 2013-02-13 | 成都印钞有限公司 | Windowing zebra crossing white watermark anti-counterfeit technology and anti-counterfeit paper made by using same |
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 |
CN106780966A (en) * | 2017-02-17 | 2017-05-31 | 深圳怡化电脑股份有限公司 | A kind of paper money discrimination method and device |
CN108711213A (en) * | 2018-03-30 | 2018-10-26 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note zebra stripes black and white block and device |
CN111915792A (en) * | 2020-05-19 | 2020-11-10 | 武汉卓目科技有限公司 | Method and device for identifying zebra crossing image-text |
-
2020
- 2020-12-31 CN CN202011635134.2A patent/CN112735021B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5093184A (en) * | 1989-06-02 | 1992-03-03 | Portals Limited | Security paper with metallic patterned elongated security element |
DE102005016807A1 (en) * | 2005-04-05 | 2006-10-12 | Adolf Wielers | Security strip for incorporation in money paper, to combat counterfeiting of banknotes, is provided and/or printed with an electronically readable code or electrically conductive chemical |
CN102926280A (en) * | 2012-10-25 | 2013-02-13 | 成都印钞有限公司 | Windowing zebra crossing white watermark anti-counterfeit technology and anti-counterfeit paper made by using same |
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 |
CN106780966A (en) * | 2017-02-17 | 2017-05-31 | 深圳怡化电脑股份有限公司 | A kind of paper money discrimination method and device |
CN108711213A (en) * | 2018-03-30 | 2018-10-26 | 深圳怡化电脑股份有限公司 | A kind of recognition methods of bank note zebra stripes black and white block and device |
CN111915792A (en) * | 2020-05-19 | 2020-11-10 | 武汉卓目科技有限公司 | Method and device for identifying zebra crossing image-text |
Also Published As
Publication number | Publication date |
---|---|
CN112735021A (en) | 2021-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN100580716C (en) | Verification method for determining areas within an image corresponding to monetary banknotes | |
JP5616958B2 (en) | Method for banknote detector device and banknote detector device | |
CN107610322B (en) | Banknote version identification method and device, electronic equipment and storage medium | |
Kumar et al. | A fast DCT based method for copy move forgery detection | |
CN107134047B (en) | White watermark detection method and device | |
CN106920318B (en) | Method and device for identifying paper money | |
Youn et al. | Efficient multi-currency classification of CIS banknotes | |
WO2023231262A1 (en) | Hoisting wire rope tension measurement method based on visual vibration frequency identification | |
CN106934922A (en) | A kind of paper currency detecting method and device | |
CN112735021B (en) | Banknote zebra crossing identification and counterfeit detection method | |
CN106296975B (en) | method and device for identifying face value of dollar paper money | |
CN107170108A (en) | One kind splicing paper money detection method and system | |
CN108510638B (en) | Paper money identification method and device | |
CN111915792B (en) | Method and device for identifying zebra crossing image-text | |
KR101232684B1 (en) | Method for detecting counterfeits of banknotes using Bayesian approach | |
CN109543554A (en) | Bill detection method, device, terminal and computer readable storage medium | |
CN107301718A (en) | A kind of image matching method and device | |
CN108629886B (en) | Method and device for detecting stain grade of paper money | |
Sooruth et al. | Automatic South African Coin Recognition Through Visual Template Matching | |
US20060115140A1 (en) | Pattern identification method and identification device | |
CN108073921B (en) | Method and device for identifying banknote information | |
CN110969757A (en) | Multi-country banknote type rapid identification technology | |
US10438436B2 (en) | Method and system for detecting staining | |
CN111627145B (en) | Method and device for identifying fine hollow image-text of image | |
JP3187698B2 (en) | Paper sheet recognition 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 |