CN107767531B - Paper money identification method and system - Google Patents
Paper money identification method and system Download PDFInfo
- Publication number
- CN107767531B CN107767531B CN201710999560.6A CN201710999560A CN107767531B CN 107767531 B CN107767531 B CN 107767531B CN 201710999560 A CN201710999560 A CN 201710999560A CN 107767531 B CN107767531 B CN 107767531B
- Authority
- CN
- China
- Prior art keywords
- segmentation
- line
- width
- initial
- cutting
- 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
Images
Landscapes
- Character Input (AREA)
- Inspection Of Paper Currency And Valuable Securities (AREA)
Abstract
The invention discloses a paper money identification method and a system, wherein the method comprises the following steps: determining an initial tangent line of the binarized crown word number image; determining the segmentation lines except the initial segmentation line: with initial tangent line as the benchmark, try the segmentation towards the opposite both sides of hat shop image in proper order, make each tangent line all be located the blank region, include: making the splitting width of each trial splitting in a preset splitting width range, and when the splitting line in the blank area cannot be obtained by the trial splitting in the preset splitting width range, determining the splitting line by taking 1/2 of the sum of the minimum width value and the maximum width value of the preset splitting width range as the actual splitting width; and respectively segmenting the crown word number image by using an initial segmentation line and segmentation lines except the initial segmentation line. The technical scheme provided by the invention can ensure the success rate and the precision of character segmentation and provide favorable conditions for paper money identification by identifying the non-blank area where the crown character of the paper money is adhered and determining the segmentation line.
Description
Technical Field
The invention relates to the field of digital image processing, in particular to a paper currency identification method and a paper currency identification system.
Background
With the development of national economy of China, various economic activities become frequent, the currency circulation of paper money on the market is increased continuously, the circulation period is shorter and shorter, and the paper money plays an important role in the life of people as a main circulation currency. However, counterfeit money and genuine money follow each other and appear in various links in the field of circulation. The inundation of counterfeit money not only threatens the property safety of people, but also seriously interferes various links of currency circulation, thereby causing the disorder of national financial order, even spreading to the political field and causing economic and social crisis. Therefore, the effective authentication of the authenticity of the paper money is an essential important work in the financial anti-counterfeit work.
The serial number of the paper money is used for recording the paper money issuing sequence, and the serial number has uniqueness, so that the serial number of each paper money is recorded when the paper money enters or exits the bank, a currency code file is established, the circulation condition of the paper money in the society can be effectively tracked by financial departments and the like, and effective supervision is carried out, thereby being beneficial to solving the important social problems like robbing money trucks, money laundering and the like. By means of the 'crown word' and 'number' on the paper money and the filing data of banks and paper money printing factories, the relevant information of the printing time, the printing place, the printing personnel, the inspection and packaging personnel and the like of any paper money can be traced. The RMB in China generally adopts 'one ticket and one number', that is, the crown word number represents the unique identity of each RMB.
When extracting the crown word number characters of the paper money, the procedures such as image data binarization processing and character cutting need to be carried out, and when cutting the crown word number characters, the prior scheme generally carries out horizontal or vertical projection on the binarized crown word number image, then sets a proper threshold value to intercept the projection of the characters, and the area larger than the threshold value is regarded as a character area, so that the crown word number characters can be cut from the image.
As shown in fig. 1, in order to obtain the crown word number projection image after the crown word number image is subjected to binarization and the crown word number image is vertically projected, it can be clearly seen that after projection, gaps exist between characters, and the gaps are the character segmentation positions. By setting a suitable threshold, the gaps can be cut apart to extract individual characters.
However, in the process of paper currency circulation, the paper currency itself is difficult to avoid the phenomena of contamination or folding angles, such as handwriting, ink, creases, and the like, which exist in and out in an ideal state of cutting the serial number characters of the paper currency, when the currency detector shoots or scans the serial number on the paper currency, the characters are adhered due to some interference, the serial number image of the paper currency cannot be well cut in a projection mode, the adhered characters may not be cut, or the stains are cut into a single character, so that the serial number identification process and precision are affected.
Disclosure of Invention
In view of the defects in the prior art, the invention provides the paper currency identification method and the paper currency identification system, which can improve the segmentation accuracy of the crown word number image and ensure the reliability and the accuracy of the crown word number identification.
In order to achieve the purpose, the invention adopts the following technical scheme:
a banknote recognition method comprising:
s01, acquiring a paper money crown word number image, carrying out binarization processing, determining an initial segmentation line of the binarized crown word number image, and enabling the initial segmentation line to be located in a blank area between characters, wherein the crown word number image comprises a first side and a second side which face opposite in the length direction;
s02, determining the cutting lines except the initial cutting line: taking the initial cutting line as a reference, and sequentially trying to cut towards the first side and the second side so as to determine a blank area and enable each cutting line to be located in the blank area;
and S03, segmenting the serial number characters in the serial number image by the initial segmentation line and the segmentation lines except the initial segmentation line respectively.
As one embodiment, the step S02 includes: and when the cutting line in the blank area cannot be obtained by trying the cutting in the preset cutting width range, 1/2 of the sum of the minimum width value and the maximum width value of the preset cutting width range is used as the actual cutting width to determine the cutting line.
As one embodiment, the step S01 includes:
s011, searching adjacent blank areas from the middle part of the crown word number image to the first side and the second side in sequence;
s012, judging whether the width of the blank area is within a preset character space range;
and S013, if the width of the blank area is within the preset character space range, taking the central position of the blank area as an initial tangent line.
As one embodiment, in step S01, if the width of the found blank area is not within the predetermined character pitch range or there is no blank area, step S03 is performed with the bisector of the number of the crown characters in the crown image.
As one embodiment, the step S02 includes:
taking the last cutting line as a reference, and trying to cut by taking the minimum width value as a cutting width;
judging whether the cutting line trying to cut is located in a blank area: if so, recording the segmentation line as a reference for next segmentation; if not, retrying to use the minimum width value to increase the preset width as a new segmentation width for segmentation until the segmentation line of the attempted segmentation is located in the blank area, and recording the segmentation line;
when the attempted cutting width is increased to be greater than or equal to the maximum width value and the attempted cutting line is not located in the blank region, 1/2 of the sum of the minimum width value and the maximum width value is used as the actual cutting width to determine the cutting line.
In one embodiment, the predetermined width is a width of one pixel.
As one embodiment, the step S02 further includes:
and judging whether the segmentation widths of two continuous adjacent characters are the minimum width value, if so, increasing the maximum width value of the preset segmentation width range of the next segmentation.
Another object of the present invention is to provide a paper money discriminating system comprising:
the device comprises an initial segmentation line determining unit, a binarization processing unit and a processing unit, wherein the initial segmentation line determining unit is used for acquiring a paper currency crown word number image, performing binarization processing, determining an initial segmentation line of the binarized crown word number image and enabling the initial segmentation line to be located in a blank area between characters, and the crown word number image comprises a first side and a second side which face opposite in length direction;
the rest of the split line determining unit is used for sequentially trying to split towards the first side and the second side by taking the initial split line as a reference so as to determine a blank area and enable each split line to be located in the blank area;
the segmentation unit is used for segmenting the serial number characters in the serial number image by the initial segmentation line and the segmentation lines except the initial segmentation line;
as one embodiment, the remaining cutting line determination unit includes a logic processing unit configured to make the cutting width of each trial cutting within a predetermined cutting width range, and determine the cutting line by using 1/2, which is the sum of the minimum width value and the maximum width value of the predetermined cutting width range, as an actual cutting width when the cutting line located in the blank region cannot be obtained by the trial cutting within the predetermined cutting width range.
As one embodiment, the initial segmentation line determining unit includes:
a searching unit: the first side and the second side are sequentially searched for adjacent blank areas from the middle part of the crown word number image;
a judging unit configured to judge whether a width of the blank area is within a predetermined character pitch range;
and the determining unit is used for taking the central position of the blank area as an initial tangent line when the width of the blank area is within the preset character spacing range.
The method firstly determines the initial segmentation line of the binarized crown word number image, then segments the crown word number image after determining the rest segmentation lines towards the two sides of the initial segmentation line in sequence, and can ensure the success rate and the segmentation precision of the character segmentation and provide favorable conditions for the banknote identification by identifying the non-blank area where the crown word number characters of the banknote are adhered and determining the segmentation lines.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic diagram of a prior art crown image and its vertically projected image;
FIG. 2 is a schematic diagram of a banknote recognition method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a method for determining an initial segmentation line according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a method for determining a dicing line other than the initial dicing line according to an embodiment of the present invention;
FIG. 5 is a schematic view of a banknote with crown-type characters stained according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of the crown word symbol of FIG. 5 after being elastically cut;
FIG. 7 is a partial schematic view of a dog-ear note in accordance with an embodiment of the present invention;
FIG. 8 is a schematic view of the crown word of FIG. 7 after being elastically cut;
fig. 9 is a schematic diagram of a banknote recognition system according to an embodiment of the present invention.
Detailed Description
In the present invention, the terms "disposed", "provided" and "connected" are to be understood in a broad sense. For example, it may be a fixed connection, a removable connection, or a unitary construction; can be a mechanical connection, or an electrical connection; may be directly connected, or indirectly connected through intervening media, or may be in internal communication between two devices, elements or components. The specific meanings of the above terms in the present invention can be understood by those of ordinary skill in the art according to specific situations.
In addition, the terms "upper", "lower", "left", "right", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings. These terms are used primarily to better describe the invention and its embodiments and are not intended to limit the indicated devices, elements or components to a particular orientation or to be constructed and operated in a particular orientation.
Moreover, some of the above terms may be used to indicate other meanings besides the orientation or positional relationship, for example, the term "on" may also be used to indicate some kind of attachment or connection relationship in some cases. The specific meanings of these terms in the present invention can be understood by those skilled in the art as appropriate.
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 2, the banknote recognition method according to the embodiment of the present invention mainly includes:
and S01, acquiring the crown word number image of the paper currency, carrying out binarization processing, determining an initial segmentation line of the binarized crown word number image, and enabling the initial segmentation line to be located in a blank area between characters, wherein the crown word number image comprises a first side and a second side which face opposite in the length direction.
For convenience of description, the following description will be made by taking the case where the serial number characters are arranged in a horizontal direction, where the first side and the second side are the left side and the right side, respectively, and it is understood that the first side and the second side may be the upper side and the lower side, respectively, in the case where the serial number characters are arranged in a vertical direction.
Fig. 3 is a schematic diagram illustrating a method for determining an initial cut line according to an embodiment of the present invention. Step S01 includes:
s011, searching adjacent blank areas from the middle part of the crown word number image to the left side and the right side in sequence;
s012, judging whether the width of the searched blank area is within a preset character spacing range W, wherein the preset character spacing range W is a reasonable interval value obtained by counting a proper amount of samples, and the interference caused by the conditions of dirt, noise and the like among 2 characters can be eliminated through the preset character spacing range W;
s013, when the width of the searched blank area is within the preset character space range W, regarding the blank area as a blank area between two characters, and taking the center position of the searched blank area as an initial tangent line.
In step S01, the blank area is searched from the middle of the crown-type image toward the left side, and when the left side does not find a suitable blank area, the blank area is searched from the middle of the crown-type image toward the right side until a suitable blank area is found. And when the widths of all the searched blank areas are not in the preset character spacing range or the blank areas do not exist, determining that the blank areas beside all the characters are stained, and taking the C bisector of the crown character image as a segmentation line to segment the crown character image, wherein C is the number of the crown character.
S02, determining the rest cutting lines (namely the cutting lines except the initial cutting line): and (3) taking the initial cutting line as a reference, and trying to cut towards the left side and the right side in sequence to determine a blank area, wherein each cutting line is positioned in the blank area.
Fig. 4 is a schematic diagram illustrating a method for determining a remaining slicing line according to an embodiment of the present invention. Here, the determination process S02 of the remaining cut lines other than the initial cut line mainly includes:
first, a predetermined segmentation width range Q is preset, wherein the predetermined segmentation width range Q is obtained by counting a proper amount of samples and is a range of a sum of a character interval (i.e., a predetermined character interval range W) and a character width, and the predetermined segmentation width range Q has a minimum width value M and a maximum width value N.
Then, the cutting line obtained by the last trial cutting is used as a reference (the initial cutting line C0 is used as a reference for the first trial cutting) to be cut towards the left side and the right side in sequence, the cutting width of each trial cutting is made to be within a preset cutting width range Q, when the cutting line in a blank area cannot be obtained by the trial cutting in the preset cutting width range Q, 1/2 of the sum of the minimum width value M and the maximum width value N of the preset cutting width range Q is used as an actual cutting width to determine the cutting line, and the cutting line is used as a reference for determining the next cutting line in the same side.
It should be noted that, when performing trial slicing with the predetermined slicing width range Q, the main steps are as follows:
firstly, trying to cut by taking a minimum width value M as a cutting width;
judging whether the cutting line trying to cut is located in a blank area: if so, recording the segmentation line as the reference of the next segmentation; otherwise, the preset width is increased by the minimum width value M to be used as a new cutting width for cutting, until the cutting line of the cutting is positioned in the blank area, and the cutting line is recorded, wherein the preset width is preferably the width of one pixel, so that any blank area cannot be missed;
when the splitting width of the trial splitting is increased to be greater than or equal to the maximum width value N and the splitting line of the trial splitting is not located in the blank region, 1/2 of the sum of the minimum width value M and the maximum width value N is used as the actual splitting width to determine the splitting line.
As shown in fig. 6 and 8, for example, when the initial splitting line C0 is determined, the trial splitting is performed toward the left side with the predetermined splitting width range Q as the splitting width of the trial splitting with reference to the initial splitting line C0, and if the splitting line falls in a blank region, the splitting is successful, and the splitting line is recorded as the first splitting line L1 on the left side, otherwise, the trial splitting is continued until the splitting line falls in the blank region or the splitting width of the trial splitting is increased to the maximum width value N. When the splitting width of the trial splitting is increased to the maximum width value N and the splitting line of the trial splitting is not located in the blank region yet, the splitting line is determined with (M + N)/2 as the actual splitting width, and the splitting line is recorded as the first splitting line L1 on the left side. Then, the trial splitting process is repeated with the first left splitting line L1 as a reference, and a second left splitting line L2 and a third left splitting line L3 … … are obtained in this order until the trial splitting is performed to the left edge of the crown image.
When the left side dividing line of the initial dividing line C0 is determined, trial division is performed to the right side with reference to the initial dividing line C0 in the same manner, and the right first dividing line R1, the right second dividing line R2, and the right third dividing line R3 … … are obtained in this order until trial division is performed to the right edge of the crown character number image.
And S03, segmenting the crown word image by the initial segmentation line C0 and the rest segmentation lines.
The determined initial splitting line C0, the first left splitting line L1, the second left splitting line L2, the third splitting line L3 … …, the first right splitting line R1, the second right splitting line R2, and the third left splitting line R3 … … are arranged from left to right, and the crown character in the crown character image is split according to the splitting lines.
In addition, since the crown word number may have a situation that a plurality of characters are all 1, the width of the character 1 itself is obviously smaller than that of other characters, which causes a plurality of continuous narrow segmentation widths (the segmentation widths are all the minimum width values M), which may cause the failure of the segmentation of the character with the thickness of not 1, that is, the width deviation accumulated for a plurality of times is too large, if the next segmentation line is determined according to the original predetermined segmentation width range Q, a blank area beside the character cannot be found certainly, which causes segmentation errors. Based on this, step S02 of this embodiment further includes a deviation rectifying step, where the deviation rectifying step first determines whether there is a situation where the segmentation widths of two consecutive adjacent characters are both the minimum width value M, and if so, increases the maximum width value N of the predetermined segmentation width range Q for the next segmentation, so as to eliminate the segmentation deviation and ensure the segmentation accuracy.
As shown in fig. 5 and 6, the banknote with stained crown-type characters and the effect schematic diagram of the separated characters show that after the stained crown-type characters are stained, each character can be accurately separated, and the influence of character adhesion caused by staining on the recognition precision is avoided. As shown in fig. 7 and 8, which are schematic diagrams illustrating the effect of the banknote with the crown word image area being blocked by the break angle and the characters thereof being cut, it can be seen that although 3 characters are blocked by the break angle, each character area can be cut after the banknote is processed according to the recognition method of the present invention.
As shown in fig. 9, the banknote recognition system according to the present invention includes:
an initial segmentation line determining unit 1, configured to acquire a banknote crown word number image, perform binarization processing, determine an initial segmentation line C0 of the binarized crown word number image, and locate the initial segmentation line C0 in a blank region between characters, where the crown word number image includes left and right sides facing opposite directions in a length direction;
the remaining cutting line determining unit 2 is configured to perform trial cutting toward the left side and the right side in sequence with the initial cutting line C0 as a reference to determine a blank area, and locate each cutting line in the blank area;
the segmentation unit 3 is used for segmenting the crown word number characters in the crown word number image by using the initial segmentation line C0 and the rest segmentation lines;
the remaining cutting line determining unit 2 includes a logic processing unit 20, which is configured to make the cutting width of each trial cutting be within the predetermined cutting width range Q, and determine the cutting line by using 1/2, which is the sum of the minimum width value M and the maximum width value N of the predetermined cutting width range Q, as the actual cutting width when the cutting line located in the blank region cannot be obtained by trial cutting in the predetermined cutting width range Q.
As one embodiment, the initial cut line determining unit 1 includes:
the search unit 11: the device is used for searching a blank area which is adjacent to the middle part of the crown word image towards the left side and the right side in sequence;
a judging unit 12, configured to judge whether the width of the found blank area is within a predetermined character pitch range W;
the determining unit 13 is configured to use the center position of the searched blank area as an initial dividing line C0 when the width of the searched blank area is within the predetermined character spacing range W.
The initial tangent line determining unit 1 first searches for a blank area from the middle of the crown-type number image toward the left side, and when the left side does not find a suitable blank area, then searches for the blank area from the middle of the crown-type number image toward the right side until a suitable blank area is found.
Meanwhile, the paper money discriminating system further includes:
and the emergency processing unit 4 is used for judging that the blank areas beside all the characters are stained when the widths of all the searched blank areas are not in the preset character interval range or the blank areas do not exist, and the emergency processing unit 4 is used for performing special processing under the condition, so that the rest of the segmentation line determining unit 2 takes the C equal division line of the crown word number image as the segmentation line, wherein C is the number of the crown word number characters, and the segmentation unit 3 is used for segmenting the crown word number image according to the C equal division line.
The remaining dividing line determining unit 2 includes:
an trial splitting unit 21 configured to try splitting with the minimum width value M as a splitting width on the basis of the previous splitting line;
a first dividing line determining unit 22 configured to determine whether the dividing line of the trial division is located in the blank region: if so, recording the segmentation line as the reference of the next segmentation; if not, retrying to increase the preset width by the minimum width value M as a new cutting width for cutting until the cutting line of the attempted cutting is positioned in the blank area, and recording the cutting line, wherein the preset width is preferably the width of one pixel;
and a second dividing line determining unit 23 for determining the dividing line with 1/2 being the sum of the minimum width value M and the maximum width value N as the actual dividing width when the attempted dividing width is increased to be greater than or equal to the maximum width value N and the attempted dividing line is not located in the blank region.
Because the crown word number may have the situation that a plurality of characters are all 1, the width of the character 1 itself is obviously smaller than that of other characters, a plurality of continuous narrow segmentation widths (the segmentation widths are all the minimum width values M) are generated, the character with the thickness not 1 is likely to fail to be segmented, namely, the width deviation accumulated for a plurality of times is overlarge, if the next segmentation line is determined according to the original preset segmentation width range Q, a blank area beside the character cannot be found certainly, and segmentation errors are caused. Based on this, the remaining segmentation line determining unit 2 of the present embodiment further includes a deviation rectifying unit 24, where the deviation rectifying unit 24 is configured to determine whether there is a situation that the segmentation widths of two consecutive adjacent characters are both the minimum width value M, and if so, appropriately increase the maximum width value N of the predetermined segmentation width range Q for the next segmentation.
The method firstly determines the initial segmentation line of the binarized crown word number image, then segments the crown word number image after determining the rest segmentation lines towards the two sides of the initial segmentation line in sequence, and can ensure the success rate and the segmentation precision of the character segmentation by identifying the non-blank area where the crown word number characters of the paper currency are adhered and determining the segmentation lines, thereby providing favorable conditions for the subsequent character identification. Because the crown word number character segmentation is not required to be carried out in a crown word number image vertical projection mode, the condition that segmentation fails due to the fact that the crown word number image is stained and the crown word number part is shielded due to folding angles and the like is reduced, even if the condition exists, the identification method can accurately segment each complete character area, the success rate and the precision of character segmentation are improved, and favorable conditions are provided for paper currency identification.
The foregoing is directed to embodiments of the present application and it is noted that numerous modifications and adaptations may be made by those skilled in the art without departing from the principles of the present application and are intended to be within the scope of the present application.
Claims (8)
1. A banknote recognition method, comprising:
s01, acquiring a paper money crown word number image, carrying out binarization processing, determining an initial segmentation line of the binarized crown word number image, and enabling the initial segmentation line to be located in a blank area between characters, wherein the crown word number image comprises a first side and a second side which face opposite in the length direction; the step S01 includes:
s011, searching adjacent blank areas from the middle part of the crown word number image to the first side and the second side in sequence;
s012, judging whether the width of the blank area is within a preset character space range;
s013, if the width of the blank area is within the preset character space range, taking the central position of the blank area as an initial tangent line;
s02, determining the cutting lines except the initial cutting line: taking the initial cutting line as a reference, and sequentially trying to cut towards the first side and the second side so as to determine a blank area and enable each cutting line to be located in the blank area;
and S03, segmenting the serial number characters in the serial number image by the initial segmentation line and the segmentation lines except the initial segmentation line.
2. The banknote recognition method according to claim 1, wherein said step S02 includes: and when the cutting line in the blank area cannot be obtained by trying the cutting in the preset cutting width range, 1/2 of the sum of the minimum width value and the maximum width value of the preset cutting width range is used as the actual cutting width to determine the cutting line.
3. The banknote discriminating method according to claim 2, wherein in step S01, if all of the blank regions found are not within the predetermined character pitch range or have no blank region, step S03 is performed with a bisector line of the number of crown characters of the crown image as a bisector line.
4. The banknote recognition method according to claim 2, wherein said step S02 includes:
taking the last cutting line as a reference, and trying to cut by taking the minimum width value as a cutting width;
judging whether the cutting line trying to cut is located in a blank area: if so, recording the segmentation line as a reference for next segmentation; if not, retrying to use the minimum width value to increase the preset width as a new segmentation width for segmentation until the segmentation line of the attempted segmentation is located in the blank area, and recording the segmentation line;
when the attempted cutting width is increased to be greater than or equal to the maximum width value and the attempted cutting line is not located in the blank region, 1/2 of the sum of the minimum width value and the maximum width value is used as the actual cutting width to determine the cutting line.
5. The banknote recognition method according to claim 4, wherein the predetermined width is a width of one pixel.
6. The banknote recognition method according to any one of claims 2 to 5, wherein said step S02 further comprises:
and judging whether the segmentation widths of two continuous adjacent characters are the minimum width value, if so, increasing the maximum width value of the preset segmentation width range of the next segmentation.
7. A banknote recognition system, comprising:
the device comprises an initial segmentation line determining unit (1) and a processing unit, wherein the initial segmentation line determining unit is used for acquiring a paper currency crown word number image, carrying out binarization processing, determining an initial segmentation line of the binarized crown word number image and enabling the initial segmentation line to be located in a blank area between characters, and the crown word number image comprises a first side and a second side which face opposite in the length direction;
the initial dividing line determining unit (1) includes:
lookup unit (11): the first side and the second side are sequentially searched for adjacent blank areas from the middle part of the crown word number image;
a judging unit (12) for judging whether the width of the blank area is within a predetermined character pitch range;
a determination unit (13) for taking the center position of the blank area as an initial tangent line when the width of the blank area is within the predetermined character pitch range;
the rest cutting line determining unit (2) is used for sequentially trying to cut towards the first side and the second side by taking the initial cutting line as a reference so as to determine a blank area and enable each cutting line to be located in the blank area;
and the segmentation unit (3) is used for segmenting the serial number characters in the serial number image by the initial segmentation line and the segmentation lines except the initial segmentation line.
8. The banknote recognition system according to claim 7, wherein the remaining segmentation line determination unit (2) includes a logic processing unit (20) for making the segmentation width of the attempted segmentation within a predetermined segmentation width range, and determining the segmentation line with 1/2, which is the sum of the minimum width value and the maximum width value of the predetermined segmentation width range, as an actual segmentation width when the segmentation line located in the blank region cannot be obtained by attempting segmentation within the predetermined segmentation width range.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710999560.6A CN107767531B (en) | 2017-10-24 | 2017-10-24 | Paper money identification method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710999560.6A CN107767531B (en) | 2017-10-24 | 2017-10-24 | Paper money identification method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107767531A CN107767531A (en) | 2018-03-06 |
CN107767531B true CN107767531B (en) | 2020-01-14 |
Family
ID=61270020
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710999560.6A Active CN107767531B (en) | 2017-10-24 | 2017-10-24 | Paper money identification method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107767531B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110837754B (en) * | 2018-08-16 | 2022-08-30 | 深圳怡化电脑股份有限公司 | Character cutting positioning method and device, computer equipment and storage medium |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000306140A (en) * | 1999-04-21 | 2000-11-02 | Oki Electric Ind Co Ltd | Conversion processor and automatic teller machine |
CN101877050B (en) * | 2009-11-10 | 2012-08-22 | 青岛海信网络科技股份有限公司 | Automatic extracting method for characters on license plate |
CN101923741B (en) * | 2010-08-11 | 2012-07-18 | 西安理工大学 | Paper currency number identification method based on currency detector |
CN106339704A (en) * | 2015-07-14 | 2017-01-18 | 富士通株式会社 | Character recognition method and character recognition equipment |
CN106296969B (en) * | 2016-08-18 | 2019-04-12 | 深圳怡化电脑股份有限公司 | The recognition methods and system of bank note |
CN106650736A (en) * | 2016-09-28 | 2017-05-10 | 浪潮软件集团有限公司 | Method for realizing Chinese character recognition system based on deep learning |
-
2017
- 2017-10-24 CN CN201710999560.6A patent/CN107767531B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN107767531A (en) | 2018-03-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104680161A (en) | Digit recognition method for identification cards | |
CN109684987B (en) | Identity verification system and method based on certificate | |
Shidore et al. | Number plate recognition for indian vehicles | |
CN101923741B (en) | Paper currency number identification method based on currency detector | |
KR101987618B1 (en) | Vehicle license plate specific system applying deep learning-based license plate image matching technique | |
CN105069900A (en) | Method and device for processing banknote information | |
EP2360619A1 (en) | Fast fingerprint searching method and fast fingerprint searching system | |
CN103034848A (en) | Identification method of form type | |
CN108269349B (en) | Paper money processing method, device, equipment and storage medium | |
CN105335741A (en) | Smudged serial number classification method and system | |
CN107767531B (en) | Paper money identification method and system | |
CN106780953B (en) | Paper money counterfeit distinguishing method and system based on double-crown-word number | |
KR102094234B1 (en) | Serial number recognition Apparatus and method for paper money | |
CN107358718A (en) | A kind of crown word number identification method, device, equipment and storage medium | |
Choi et al. | Localizing slab identification numbers in factory scene images | |
US20150379338A1 (en) | Apparatus and Method for Recognizing Media and Financial Device | |
CN107170109B (en) | Paper money watermark authenticity distinguishing method and device | |
CN112509209A (en) | Paper currency magnetic detection method and device, currency detection equipment and readable storage medium | |
CN105787491A (en) | Feature extraction method of fingerprint image | |
CN112070020B (en) | Method and device for counting number of people entering well, electronic equipment and readable storage medium | |
CN109410420B (en) | Image detection method and device and bill discriminator | |
CN110276353B (en) | Method and device for segmenting prefix number characters, readable storage medium and terminal equipment | |
CN102364526A (en) | Method, device and system for discriminating bank magnetic card | |
Kwon et al. | Recognition based verification for the machine readable travel documents | |
CN106504405B (en) | A kind of method and device identifying bank note |
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 |