KR101470748B1 - Character recognition method, character recognition apparatus and financial apparatus - Google Patents

Abstract

The present invention relates to a character recognizing method for recognizing characters of a medium. The character recognizing method includes the steps of: obtaining a character image of an individual character from a medium; and determining a character by reading, in a stepwise manner, the character image according to a hierarchical structure in which a group of certain characters is hierarchically classified into multiple groups including superordinate and subordinate groups.

Description

Technical Field [0001] The present invention relates to a character recognition method, a character recognition apparatus and a financial apparatus,

A character recognition method, and a financial instrument.

Financial equipment is a device that processes financial business such as inputting or withdrawing a medium such as banknotes or checks. It is important to accurately recognize the characters included in the medium in order to enhance the reliability of the financial business.

In order to improve the accuracy of character recognition, an optical character reading method is widely used in recent years. However, in the case of the optical character reading system, when the quality of the scanned image is good, the accuracy of character recognition is good, but when the quality of the scanned image is poor, the accuracy of character recognition is low. Accordingly, there is a need for a character recognition method capable of accurately and quickly recognizing characters on a medium even when the quality of the scanned image is poor.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and it is an object of the present invention to provide a character recognition method, a character recognition apparatus, and a financial instrument. The character recognition method, the character recognition apparatus, and the financial apparatus are not limited to the above-described technical problems, and other technical problems may exist.

A character recognition method according to one aspect includes: obtaining a character image of an individual character from the medium; And determining a character by stepwise reading the character image according to a hierarchical structure hierarchically classifying a set of predetermined characters into a plurality of groups composed of upper groups and lower groups.

According to another aspect of the present invention, there is provided a character recognition apparatus comprising: an image obtaining unit obtaining a character image of an individual character from a medium; And a character determination unit for determining the individual character from a character image of the individual character; Wherein the character determination unit determines a predetermined number of alphabetic characters of the same number of the different media in a predetermined number of characters when performing determination of the individual characters over a predetermined number of times for different media.

According to another aspect of the present invention, there is provided a character recognition apparatus comprising: an image obtaining unit obtaining a character image of an individual character from a first bank bill; And a character determination unit for determining an individual character from the character image of the individual character, wherein when the alphabetical character of the second station serial number is substituted for the numeric character digit of the first station serial number of the first national bills, The alphabetical character of the binary serial number is determined as a number in the set of numbers including 0 to 9 or a number in the set containing the numbers.

According to another aspect of the present invention, there is provided a financial device comprising: a media input / output unit for inputting / withdrawing a medium including characters; Obtaining a character image of an individual character from the medium and determining the character by reading the character image according to a hierarchical structure hierarchically classifying a set of predetermined characters into a plurality of groups composed of upper groups and lower groups A character recognition device; And a controller for controlling operations of the media input / output unit and the character recognition apparatus.

A character recognition method according to another aspect of the present invention provides a computer-readable recording medium having recorded thereon a program for execution on a computer.

According to the above, characters included in the medium can be recognized accurately and quickly.

1 is a block diagram illustrating a financial device according to an embodiment of the present invention.
2 is a block diagram illustrating a character recognition apparatus according to an embodiment of the present invention.
3 is a diagram for explaining a method for a character recognition apparatus to determine a character by reading a character according to a hierarchical structure according to an embodiment of the present invention.
4 is a diagram for explaining a character image according to an embodiment of the present invention.
5 is a diagram for explaining an example of determining a character of a medium using a hierarchical structure in a character recognition apparatus according to an embodiment of the present invention.
6 is a flowchart illustrating a character recognition method for determining a character of a medium according to an embodiment of the present invention.
7 is a flowchart illustrating a character recognition method for determining a character of a medium according to another embodiment of the present invention.

Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. It should be noted that, in adding reference numerals to the constituent elements of the drawings, the same constituent elements are denoted by the same reference numerals whenever possible, even if they are shown in different drawings. In the following description of the embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the difference that the embodiments of the present invention are not conclusive.

In describing the components of the embodiment of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are intended to distinguish the constituent elements from other constituent elements, and the terms do not limit the nature, order or order of the constituent elements. When a component is described as being "connected", "coupled", or "connected" to another component, the component may be directly connected or connected to the other component, Quot; may be "connected," "coupled," or "connected. &Quot;

The financial device according to an embodiment of the present invention may be a financial device that obtains various media such as banknotes, securities, geographical coins, coins, gift certificates and the like and processes and / or withdraws money such as deposits, And the like, and performs a financial task to perform the media processing such as the processing such as the release. Examples of such financial instruments include an automated teller machine (ATM) such as a cash dispenser (CD), a cash recycling machine, and the like. However, the financial instrument is not limited to the above-described example, and may be a device for automating financial business such as FIS (Financial Information System).

Hereinafter, an embodiment of the present invention will be described on the assumption that the financial instrument is a financial automatic instrument. However, this assumption is for convenience of explanation, and the technical idea of the present invention is not limited to the financial automation equipment.

1 is a block diagram illustrating a financial device according to an embodiment of the present invention. 1, the financial instrument 100 includes a media input / output unit 110, a control unit 120, and a character recognition apparatus 200. [ The financial instrument 100 may further include a storage unit 130, a user interface unit 140, or a memory 150.

The medium input / output unit 110 draws in or draws out the medium. For example, the medium may be, but is not limited to, a bill, a check, a paper clip, or a gift certificate. The media input / output unit 110 may further include a sensor for determining whether or not the medium has been inserted or withdrawn.

The character recognition apparatus 200 recognizes the medium and determines a character. For example, the character recognition apparatus 200 can determine a character using an optical character reading system. The character recognition apparatus 200 can acquire an image of the medium in such a manner that the light emitted from the light source is reflected on the medium and the reflected or transmitted light is received on the medium. The character recognition apparatus 200 determines the character included in the medium by using the image of the obtained medium. See FIG. 2 for a detailed description related to this.

The control unit 120 controls the overall operation of the financial instrument 100. [ That is, the control unit 120 controls the media input / output unit 110 to draw in or out the medium, or controls the character recognition device 200 to recognize the printed characters on the medium, And controls the user interface unit 140 to receive data such as a command or information from the user or to receive data from the user via the media input / To display the information.

For example, when the control unit 120 detects that the media input to the media input / output unit 110 is being transferred from the media input / output unit 110 to the character recognition device 200, It is possible to control to determine printed characters.

The controller 120 according to the present embodiment may correspond to at least one processor or may include at least one processor. Accordingly, the control unit 120 may be operated in a form included in another hardware device such as a microprocessor or a general-purpose computer system.

The storage unit 130 stores the medium. For example, the storage unit 130 may receive and store the media input from the media input / output unit 110, or may transfer the stored media to the media input / output unit 110 for fetching.

The user interface unit 140 acquires input information from a user and displays output information to a user. For example, the user interface unit 140 may display a character recognized by the character recognition apparatus 200 or may indicate that a character recognition error occurs when the character recognition apparatus 200 fails to recognize the character . If necessary, the user interface unit 140 may receive character information for designating characters that fail to recognize characters from the user.

The user interface unit 140 may include input / output devices such as a display panel, a touch screen, a monitor, a keyboard, and a speaker, and a software module for driving them.

The memory 150 stores information necessary for character recognition. For example, the memory 150 may include a hierarchical structure in which a predetermined set of characters is hierarchically classified into a plurality of groups composed of upper groups and lower groups, conditions for reading character images step by step according to a hierarchical structure , And a template necessary for character reading. However, such a template and the like may be stored in the memory 150 in the form of a program included in a program for reading characters, rather than being stored alone in the memory. In addition, the memory 150 may store characters recognized by the character recognition apparatus 200. [

The memory 150 may be a hard disk drive (HDD), a hard disk drive (HDD), a hard disk drive (HDD) The present invention can be implemented by a ROM, a RAM, a flash memory, a memory card, and a solid state drive (SDD).

2 is a block diagram illustrating a character recognition apparatus according to an embodiment of the present invention. Referring to FIG. 2, the character recognition apparatus 200 is a device for recognizing characters included in a medium, and may include an image acquisition unit 210 and a character determination unit 220. Since the contents described in relation to the character recognition apparatus 200 in FIG. 1 are also applicable to FIG. 2, duplicated description thereof will be omitted.

The character recognition apparatus 200 according to the present embodiment may correspond to at least one processor or may include at least one processor. Accordingly, the character recognition apparatus 200 can be operated in a form included in another hardware apparatus such as a microprocessor or a general-purpose computer system.

The image acquisition unit 210 and the character determination unit 220 may be driven by one processor included in the character recognition apparatus 200 or may be driven by a separate processor.

The image acquiring unit 210 acquires character images of individual characters from the medium. See FIG. 4 for an example of a character image of an individual character. The image acquiring unit 210 may acquire an image of the medium using light. After acquiring the image of the medium, the image acquiring unit 210 may perform various image processing to acquire a character image of an individual character. Accordingly, the image obtaining unit 210 may include a light emitting unit including a light source, a light receiving unit, and an image processing apparatus. See FIG. 4 for a detailed description related to this.

The character determination unit 220 determines individual characters from the character images of the individual characters acquired by the image acquisition unit 210. The character determination unit 220 determines the character by reading it according to the hierarchical structure. The hierarchical structure is a hierarchical classification of a set of predetermined characters into a plurality of groups composed of upper groups and lower groups. Here, the characters included in the upper group include characters included in the lower groups of the upper group.

According to one embodiment, the hierarchical structure may classify a predetermined set of characters into a plurality of groups composed of upper groups and lower groups based on the similarity of shapes among characters included in the set of predetermined characters .

For example, the set of predetermined characters may be a set of numbers 0 to 9, a set of alphabets A to Z, and the like. The character may include all the visual symbol systems such as Hangul, alphabet, kanji, roman, and numeral, so the set of predetermined characters is not limited to the above example.

According to the above example, characters can be read step by step according to a hierarchical structure hierarchically classified into a plurality of groups composed of upper groups and lower groups. This hierarchical structure can be obtained by classifying the sets of numbers 0 to 9 into a plurality of groups based on the similarity of the shapes of the respective numbers between the numbers 0 to 9. [

In stepwise reading the character image according to the hierarchical structure, the character determination unit 220 determines the number of character pixels, the number of background pixels, or the number of character pixels in a predetermined position of the character image, The ratio of the number of pixels can be compared with the reference value, and the character image can be read based on the comparison result.

According to one embodiment, the character determination unit 220 may determine a character using at least one template in determining a character. For example, the character determination unit 220 may position the template at a predetermined position on the character image in accordance with a control signal from the control unit 120, Characters can be determined by confirming the adjacency in the longitudinal and lateral directions. The species and transversal adjacency between character pixels can be ascertained by the number of background pixels between two sets of character pixels at different positions.

At this time, the character determination unit 220 can use a plurality of templates of various sizes to read characters included in a predetermined set of characters. The character determination unit 220 may use various templates stored in the memory 150 according to the size and type of characters.

In addition, the character determination unit 220 may hierarchically apply a plurality of templates according to a hierarchical structure when reading a character.

According to the embodiments described above, the character recognition apparatus 200 can classify character images in a hierarchical manner according to a hierarchical structure hierarchically classifying a set of predetermined characters into a plurality of groups composed of upper groups and lower groups The characters included in the medium can be accurately and quickly recognized.

3 is a diagram for explaining a method for a character recognition apparatus to determine a character by reading a character according to a hierarchical structure according to an embodiment of the present invention. The method of reading characters shown in Fig. 3 can be processed in a time-series manner in the character recognition apparatus 200 shown in Figs. Therefore, even if the contents are omitted in the following description, it can be understood that the description described above with respect to the character recognition apparatus 200 shown in Figs. 1 and 2 also applies to Fig.

After the character image of the individual character is obtained by the image obtaining unit 210 and the character image is input from the image obtaining unit 210 to the character determining unit 220, the character determining unit 220 determines the character according to the hierarchical structure Read it step by step. It is to be understood by those skilled in the art that the hierarchical structure disclosed in FIG. 3 is merely an embodiment, and various hierarchical structures other than the hierarchical structure can be applied.

The hierarchical structure hierarchically classifies a set of predetermined characters into a plurality of groups composed of upper groups and lower groups, and the characters included in the upper group include characters included in the lower groups of the upper group . The hierarchical structure may be generated in advance in the external device, and then stored in the memory 150 and used. Hereinafter, for the sake of convenience of explanation, it is assumed that a hierarchical structure is generated by one processor as an external device. The processor can create a hierarchy in the following manner.

First, the processor 1 determines a condition 1 as a reference for grouping a predetermined set of characters into highest-level groups. For example, condition 1 may be determined based on the similarity of the forms between characters included in the set of predetermined characters.

The processor classifies a set of predetermined characters according to condition 1. [ Accordingly, the characters satisfying the condition 1 can be grouped into the group A, and the characters that do not satisfy the condition 1 can be grouped into the group B.

Next, the characters included in each of the top groups may be grouped into subgroups according to a predetermined criterion. For example, based on the type of characters included in each group, each of the conditions for grouping the characters included in each group into lower groups is determined, and the characters included in each group are classified do.

According to FIG. 3, the characters of the group A are classified according to the condition 2. The characters that satisfy the condition 2 among the characters included in the group A are the group a, and the characters that do not satisfy the condition 2 are the group b . The characters in the B group are classified according to the condition 3, and the characters included in the B group are grouped into the c group and the characters that do not satisfy the condition 3 are grouped into the d group.

Further, the subgroups may be further subdivided according to predetermined criteria. Referring to FIG. 3, the characters included in the c-th group can be classified according to the condition 4, and the c-th group can be further subdivided into the first group and the second group. Accordingly, among the characters included in the c-th group, the characters satisfying the condition 4 can be grouped into the first group, and the characters that do not satisfy the condition 4 can be grouped into the second group.

In accordance with one embodiment, when creating a hierarchy for a set of numbers 0 through 9, the hierarchy is a hierarchical structure in which the numbers of numerals 0 through 9 are similar in shape to a plurality of groups of higher groups and subgroups They can be hierarchically grouped.

First, a set of numbers 0 to 9 can be divided into a group A including numbers 1 and 7 and a group B including numbers 0, 2, 3, 4, 5, 6, 8, have. Next, when the group A of numbers 1 and 7 and the group B of numbers 0, 2, 3, 4, 5, 6, 8, and 9 are further subdivided into subgroups, group A includes the number 1 Group B containing the numbers a, b, c and d, the group B containing the numbers a, b, c, d, Can be divided. The c-th group of numbers 2, 5, and 6 may be grouped into a lower group of the second group including the first group including the number 2 and the numbers 5 and 6.

The character determination unit 220 determines the character by stepwise reading the input character image using the hierarchical structure as described above. According to the hierarchical structure of FIG. 3, the character determination unit 220 determines whether the input character image satisfies the condition 1 by applying the first template to the input character image. If the inputted character image satisfies the condition 1, it is judged to belong to the group A, and if the character image does not satisfy the condition 1, it corresponds to the group B.

Next, if the character image corresponds to the group A, the character determination unit 220 determines whether the input character image satisfies the condition 2 by applying the second template to the character image. If the character image corresponds to the B group, the character determination unit 220 determines whether the input character image satisfies the condition 3 by applying the third template to the character image.

Referring to FIG. 3, if the character image corresponds to the group A and the condition 2 is satisfied, it can be determined that the character image corresponds to the group a. Alternatively, when the character image corresponds to the B group and the condition 3 is satisfied, the character determination unit 220 applies the template 4 to the character image to determine whether the character image satisfies the condition 4. At this time, the character determination unit 220 determines that the character image belongs to the first group if the character image satisfies the condition 4, and belongs to the second group if the character image does not satisfy the condition 4.

In this way, the character determination unit 220 can apply the template to each layer of the hierarchical structure, and can determine the character by hierarchically reading the input character image according to the hierarchical structure. The character determination unit 220 compares the number of character pixels at a predetermined position of the character image, the number of background pixels, or the ratio of the number of character pixels and the number of background pixels to the reference value for each step of the hierarchical structure , And based on the comparison result, the character image can be read step by step as shown in Fig.

For example, the condition 1 to the condition 4 may be whether or not the number of character pixels within a predetermined range at a predetermined position is equal to or greater than a reference value. Alternatively, the conditions 1 to 4 may be whether the ratio of the number of character pixels to the number of background pixels is equal to or greater than a reference value.

The character determination unit 220 can read the character image using the hierarchical structure of FIG. 3, and can determine the group to which the character image belongs among the groups shown in FIG. 3 based on the read result. The character determination unit 220 may determine the character of the character image as one of the characters included in the determined group. According to one embodiment, the character determination unit 220 can determine a character as a single character using at least one template in the group.

Accordingly, the character recognition apparatus 200 does not determine the character by comparing all the templates applied to the character image of each individual character in determining the character by reading the character image. According to one embodiment, the group to which the character belongs is narrowed down to a lower group according to whether or not pixels of a partial region of a predetermined position satisfy predetermined conditions according to a hierarchical structure composed of upper groups and lower groups The character to be compared can be greatly reduced in determining each character, and the character of the character image can be quickly recognized.

4 is a diagram for explaining a character image according to an embodiment of the present invention. The contents described below can be applied to the character recognition apparatus 200 shown in FIGS. 1 and 2. FIG.

Referring to Figure 4, a character image of an individual character is shown. The character image in Fig. 4 is composed of background portions except for the character portion and the character. That is, the character image is composed of the character pixels occupied by the character portion and the background pixels occupied by the background portion. The background pixels may be determined by pixels having a pixel value equal to or greater than a predetermined pixel value among the pixels constituting the character image, and the character pixels may be determined as pixels having a pixel value equal to or less than a predetermined pixel value. At this time, the pixel value may be a brightness value or a gray value of the pixel. As an example, a predetermined pixel value for separating background pixels may be 200, and a predetermined pixel value for separating character pixels may be 30.

For example, the character recognition apparatus 200 may classify a predetermined set of characters into a set of characters based on the similarity of the positions and configurations of the character pixels and the background pixels in the character image of each of the characters included in the predetermined character set, And a hierarchical grouping hierarchically grouped into subgroups. The character recognition apparatus 200 reads character images of the medium input to the character recognition apparatus 200 step by step using the hierarchical structure.

According to one embodiment, the character recognition apparatus 200 can acquire a character image in the following manner.

The image obtaining unit 210 obtains a character image of each of the characters included in the medium. In order to obtain a character image of an individual character, the image obtaining section 210 first acquires an image of the medium. At this time, the image obtaining unit 210 may obtain only a specific portion of the image, rather than an entire image of the medium.

The position of the image of the medium that the image acquisition unit 210 should acquire from the medium may be predetermined. For example, the memory 150 may store coordinate information regarding the position of the image of the medium to be acquired. In addition, since the size of the medium varies according to the types of various media, the memory 150 may store coordinate information for various types of media.

Accordingly, the image obtaining unit 210 obtains the image of the medium with reference to the coordinate information stored in the memory 150. [

Next, the image obtaining unit 210 obtains a character string image from the image of the obtained medium. The image obtaining unit 210 may perform image processing to remove a margin surrounding the character string from the image of the obtained medium to obtain a character string image. The string image contains at least one character.

After obtaining the character string image, the image obtaining unit 210 obtains the character character image of the individual characters in the obtained character string image. The size of the character image of individual characters can be set to a predetermined size. At this time, the predetermined size may be changed according to the type of the medium recognized by the character recognition apparatus 200. For example, when the character recognition apparatus 200 recognizes the serial number of the Korean won bill, the size of the character image may be determined as the font size of the serial number of the Korean won bill.

The image obtaining unit 210 may divide a character string image into character images of a predetermined size for each character using a margin between characters. The image obtaining unit 210 may display each character image as a predetermined size image composed of a plurality of pixels. For example, the image obtaining unit 210 may perform image processing such as enlarging or reducing a character image of an individual character to a predetermined size, or dividing a character image into a plurality of pixels.

Accordingly, the image obtaining unit 210 can obtain a character image as many as the number of characters included in the character image.

5 is a diagram for explaining an example of determining a character of a medium using a hierarchical structure in a character recognition apparatus according to an embodiment of the present invention. Hereinafter, the character recognition apparatus 200 recognizes the serial number 10 of the banknote shown in FIG. 5 as an example.

When banknotes are inserted into the character recognition apparatus 200, the image obtaining unit 210 obtains the image of the portion where the serial number 10 of the banknote is printed. After obtaining the image of the printed portion of the serial number 10 of the banknote, the image obtaining unit 210 cuts up and down, left and right margin portions in the obtained image, and obtains a character string image including the character string of the serial number 10 do.

The image acquiring unit 210 acquires character images of individual characters by dividing each character of the serial number 10 in the acquired character image. In the case of the bill shown in Fig. 5, a total of ten character images are obtained. A total of 10 character images acquired by the image acquisition unit 210 are sequentially transmitted to the character determination unit 220 and read out.

In the case of the serial number 10 of the banknote, the type of the character of each digit of the serial number 10 is determined according to the kind of the banknote and the denomination. In the case of the banknote shown in FIG. 5, the alphabet is placed in the first and second digits of the serial number 10, and the digits are placed in the remaining digits.

Accordingly, the character determination unit 220 determines that the first and second digits of the serial number 10 have a hierarchical structure in which a set of alphabets A through Z is hierarchically classified into a plurality of groups composed of upper groups and lower groups The input character image is read step by step and the input character image is read step by step using a hierarchical structure in which the groups of numbers 0 to 9 are hierarchically classified into a plurality of groups.

When recognizing a character image of a number in the character recognition apparatus 200, a hierarchical structure of a set of numbers is used. When recognizing a character image of an alphabet, a character image is read stepwise using a hierarchical structure of an alphabet set.

If the character recognition apparatus 200 reads a character image of an alphabet using a hierarchical structure of a number set, it can be recognized as a specific number by reading a character image of a specific alphabet.

For example, when the character recognition apparatus 200 reads the character image of the alphabet E using the hierarchical structure of the numeric set, it can be recognized as the number 2. As another example, if the character images of the alphabet E, C, or K are read using the hierarchical structure of the numeric set, it can be recognized as any of the numbers 5, 6, As another example, if a character image of the alphabet A, B, D, H, or J is read using the hierarchical structure of the numeric set, it is classified into groups including numbers 3, 4, 8, and 9, 8 and < RTI ID = 0.0 > 9, < / RTI >

Conversely, if the character image of the numeral 8 is read step by step using the hierarchical structure of the alphabet set, it can be recognized as the alphabet B.

According to one embodiment, the character recognition apparatus 200 that recognizes the serial number of the Chinese banknote will be described. First, the character recognition apparatus 200 acquires the character string image of the serial number from the Chinese banknote, Obtain a character image of the character located by each digit of the serial number.

The character recognition apparatus 200 recognizes a character image according to a hierarchical structure hierarchically classifying a set of predetermined characters assigned to each digit of the serial number of Chinese banknotes into a plurality of groups composed of upper groups and lower groups The characters are read out step by step.

In the case of Chinese banknotes, the last digit of the serial number is a number from 0 to 9. On the other hand, in the case of Korean notes, the letters A to Z are located at the end of the serial number.

Accordingly, if the Chinese banknote edited by replacing the serial number of the Korean banknote with the serial number of the Chinese banknote is recognized by the character recognition apparatus 200 that recognizes the Chinese banknote, the character recognition apparatus 200 recognizes the Chinese banknote We will recognize the last digit of the serial number as a non-alphanumeric character.

At this time, if a character image of a specific alphabet is read using a hierarchical structure for recognizing the serial number of the Chinese banknote, it can be recognized as a specific number. For example, if the last digit of the serial number of the Chinese note is alphabet E, it may be determined to be the number 2.

In the above-described embodiment, the image of the serial number of the Korean banknote replacing the serial number of the Chinese banknote can be utilized, for example, the character image of FIG. 4 described above. In this case, the font and the font size of the serial number in the character image may be the font and the font size used in the Korean currency.

In the above embodiment, the alphabetical character of the serial number of the Korean banknote is replaced with the numeric character of the serial number of the Chinese banknote. However, the present invention is not limited to this, Can be replaced with a predetermined character of the second station serial number.

Accordingly, when the alphabetical character of the serial number of the second station is substituted for the digit number of the serial number of the first station of the first country banknote, the image obtaining unit 210 obtains the character image of the individual character from the first country banknote In determining the individual character from the character image of the individual character in the character determination unit 220, an alphabetical character of the second station serial number is included in the set of numbers including 0 to 9, Can be determined by the number in the set.

As in the above-described examples, the character determination unit 220 can perform determination of a character more than a predetermined number of times for different media. In the case of performing determination of characters more than a predetermined number of times for different media, The predetermined number of alphabetic characters of the same number in the set of predetermined numeric characters can be determined. Here, the set of predetermined numeric characters may be any of upper groups or lower groups.

For example, the same number of digits of different media can be the last digit of the serial number, and if the predetermined number of English characters of the last digit is English E, K or C, then the set of predetermined numeric characters is 5, 6, 2 < / RTI > As another example, if the predetermined alphabet of the last digit is English E, the predetermined set of characters may be two alphabets. As another example, if the predetermined alphabet of the last digit is English A, B, D, H, or J, the predetermined set of numeric characters may be 3, 4, 8, and 9. As another example, if the predetermined number of English characters in the last digit is L, the set of predetermined numeric characters may be 1 and 7.

In this way, the character recognition apparatus 200 determines the character of the medium by applying a predetermined hierarchical structure for each type of medium, the type of the character, and the number of digits of the character.

6 is a flowchart illustrating a character recognition method for determining a character of a medium according to an embodiment of the present invention. Referring to FIG. 6, the method illustrated in FIG. 6 comprises steps that are processed in a time-series manner in the character recognition apparatus 200 shown in FIG. 1 to FIG. Therefore, even if the contents are omitted in the following description, it can be understood that the contents described above with respect to the character recognition apparatus 200 shown in Figs. 1 and 2 also apply to the method shown in Fig.

In operation 610, the image obtaining unit 210 obtains a character image of an individual character. For example, the image acquiring unit 210 may acquire a character image using light.

In operation 620, the character determination unit 220 determines the character by reading the character image stepwise according to the hierarchical structure. The hierarchical structure is a hierarchical classification of a set of predetermined characters into a plurality of groups composed of upper groups and lower groups, and may be a set of numbers 0 to 9 or a set of alphabets A to Z, and the like. The character determination unit 220 may determine a character of a character image using at least one template according to the type of characters included in a predetermined set of characters.

7 is a flowchart illustrating a character recognition method for determining a character of a medium according to another embodiment of the present invention. Referring to FIG. 7, the method illustrated in FIG. 7 comprises steps that are processed in a time-series manner in the character recognition apparatus 200 shown in FIG. 1 to FIG. Therefore, even if the contents are omitted in the following description, it can be understood that the above-described contents of the character recognition apparatus 200 shown in Figs. 1 and 2 also apply to the method described in Fig.

In operation 710, the character recognition apparatus 200 receives the medium input through the medium input / output unit 110.

In operation 720, the image obtaining unit 210 obtains an image of the medium. The image acquiring unit 210 may acquire only an image of a specific portion rather than an entire image of the medium.

In operation 730, the image acquisition unit 210 acquires a character string image. For example, the image obtaining unit 210 may perform image processing to remove margins to obtain a character string image including at least one character.

In operation 740, the image acquisition unit 210 acquires character images of individual characters. For example, the image obtaining unit 210 may divide a character image into character images of a predetermined size for each character through image processing.

In step 750, the character determination unit 220 determines the character by reading the character image step by step according to the hierarchical structure. For example, if the group to which the character image belongs is determined among the plurality of groups according to the reading result, the character determining unit 220 can determine the character of the character image as one of the characters included in the determined group.

In step 760, the character determination unit 220 determines whether or not the character of the input character image has been recognized, and transmits the determination result to the control unit 120. If the character of the character image is determined to be one of the predetermined set of characters according to the result of reading the character in step 750 and the character of the character image is not determined in step 770,

In operation 770, the control unit 120 processes the input medium according to the recognized result of the character recognition apparatus 200.

In operation 780, the user interface unit 140 indicates that a character recognition error has occurred in the input medium.

According to the embodiments of the present invention, the character recognition apparatus 200 may compare some pixels at a predetermined position using a hierarchical structure composed of upper groups and lower groups, The character of the image can be quickly recognized.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments. That is, within the scope of the present invention, all of the components may be selectively coupled to one or more of them. In addition, although all of the components may be implemented as one independent hardware, some or all of the components may be selectively combined to perform a part or all of the functions in one or a plurality of hardware. As shown in FIG. The codes and code segments constituting the computer program may be easily deduced by those skilled in the art. Such a computer program can be stored in a computer-readable storage medium, readable and executed by a computer, thereby realizing an embodiment of the present invention. As the storage medium of the computer program, a magnetic recording medium, an optical recording medium, a carrier wave medium, or the like may be included.

Furthermore, the terms "comprises", "comprising", or "having" described above mean that a component can be implanted unless otherwise specifically stated, But should be construed as including other elements. All terms, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined. Commonly used terms, such as predefined terms, should be interpreted to be consistent with the contextual meanings of the related art, and are not to be construed as ideal or overly formal, unless expressly defined to the contrary.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

200 ... character recognition device
210 ... image acquisition unit
220 ... character determination section

Claims (18)

1. A character recognition method for recognizing a character of a medium,
Obtaining a character image of an individual character from the medium; And
The number of character pixels in a predetermined area on the character image, the number of background pixels in each layer, and the number of background pixels in a hierarchical structure hierarchically classified into a plurality of groups, And comparing the ratio of the number of character pixels to the number of background pixels with a reference value to stepwise read the character image to determine a character. The method according to claim 1,
The step of determining
Wherein at least one template is located at a predetermined position and a character is determined by confirming the proximity between the character pixels included in the template. The method according to claim 1,
Wherein the hierarchical structure classifies the set of predetermined characters into the plurality of groups on the basis of a similarity of shapes among characters included in the set of predetermined characters. The method according to claim 1,
The step of determining
Determining a group to which the character image belongs among the plurality of groups based on the reading result; And
And determining the character of the character image as one of the characters included in the determined group. The method according to claim 1,
The hierarchical structure is a hierarchical classification of numbers 0 to 9 into a plurality of groups composed of upper groups and lower groups,
And recognizing the character image as a specific number when a character image of a specific alphabet is read using the hierarchical structure. The method according to claim 1,
The hierarchical structure is a hierarchical classification of numbers 0 to 9 into a plurality of groups composed of upper groups and lower groups,
And if the alphabet E is read using the hierarchical structure, a number 2 is determined. The method according to claim 1,
The hierarchical structure is a hierarchical classification of numbers 0 to 9 into a plurality of groups composed of upper groups and lower groups,
And if the alphabet E, C, or K is read using the hierarchical structure, it is recognized as any one of numbers 5, 6, and 2. [ The method according to claim 1,
The hierarchical structure is a hierarchical classification of alphabets A through Z into a plurality of groups composed of upper groups and lower groups,
And if the number 8 is read using the hierarchical structure, it is recognized as an alphabet B. The method according to claim 1,
Wherein the determining step determines the character using at least one template according to a type of characters included in the predetermined set of characters. A computer-readable recording medium storing a program for causing a computer to execute the method according to any one of claims 1 to 9. An image obtaining unit obtaining character images of individual characters from the medium; And
And a character determination unit for determining the individual character from a character image of the individual character,
Wherein the character determination unit determines a predetermined number of alphabetic characters of the same number of different media in a set of predetermined numeric characters when the determination of the individual characters is performed more than a predetermined number of times for different media. 12. The method of claim 11,
Wherein when the same number of digits of the different media is the last digit of the serial number and the predetermined English character is English E, K or C, the set of predetermined numeric characters is 5, 6 and 2. 12. The method of claim 11,
Wherein when the same number of digits of the different media is the last digit of the serial number and the predetermined English character is English E, the set of predetermined characters is a 2-character recognition device. 12. The method of claim 11,
If the same number of digits of the different mediums is the last digit of the serial number and the predetermined alphabet letter is A, B, D, H, or J, then the predetermined set of numeric characters is 3, 4, Device. 12. The method of claim 11,
Wherein when the same number of digits of the different media is the last digit of the serial number and the predetermined English character is English L, the set of predetermined numeric characters is 1 and 7, respectively. An image obtaining unit for obtaining a character image of individual characters from the first bank bill; And
A character determination unit for determining individual characters from character images of individual characters,
, ≪ / RTI &
When the alphabetical character of the second station serial number is replaced with the numeric character digit of the first station serial number of the first country banknote, the alphabetical character of the second station serial number is a set of numbers including 0 to 9 The number being determined by a number in the set that includes numbers or numbers. 17. The method of claim 16,
The character image of the alphabetical character of the second station serial number replaced with the numeric character digit of the first station serial number of the first country banknote includes a background pixel having a pixel value of 200 or more and a character pixel having a pixel value of 30 or less The character recognition apparatus comprising: A medium input / output unit for inputting or outputting a medium including characters;
A character image of an individual character is obtained from the medium, and a predetermined character group is hierarchically classified into a plurality of groups composed of upper groups and lower groups, A character recognition device that compares one of the number of character pixels, the number of background pixels, or the ratio of the number of character pixels and the number of background pixels to a reference value to determine the character by reading the character image; And
And a controller for controlling operations of the media input / output unit and the character recognition apparatus.

Images