JP2007065736A - Character recognition device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To notify an operator of warning when the possibility of erroneous recognition is high even if the number in a digit whose erroneous recognition is not checkable based on the configuration of a numeral sequence. <P>SOLUTION: The image of a recording medium is read with a reading device, and a numeral string described in a recording medium is character-recognized from the image read from the reading device, and whether or not possibility that the number in a predetermined digit in the recognized digit sequence has been erroneously recognized is high is decided based on a preliminarily set validity value, and when it is determined that possibility that a number in a predetermined digit has been erroneously recognized is high by the determination, warning to confirm the number in the predetermined digit as well as the recognized numeral sequence is displayed. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a character recognition device and a character recognition method for recognizing a character string by reading an image of a recording medium such as an application form or ID card in which a character string of a predetermined format such as identification information is described. .

  Conventionally, for example, in an information management system that manages various types of information in association with identification information, information associated with the identification information is stored in a database. For this reason, in the information management system, when specific information is extracted from the database, desired information is searched and extracted using the identification information as described above as a key. Further, in the information management system as described above, it is necessary to input identification information in order to extract desired information. In particular, an information management system in which identification information input work frequently occurs has been proposed to have a character recognition device in order to make the identification information input work more efficient. In an information management system having a character recognition device for recognizing such identification information, an image of a recording medium such as an application form or ID card in which the identification information is written is read, and the identification information is recognized from the image. .

  The identification information recognized by the character recognition device as described above is generally confirmed manually by human eyes. Further, in order to mechanically check for erroneous recognition by the character recognition device as described above, a number string organized based on a predetermined rule may be adopted as identification information. In such a case, it is possible to check the validity of the numeric string as the recognition result based on the numeric string system. For example, if a check digit is set for a predetermined digit in a numeric string, it is possible to check the validity of the numeric string using the check digit.

  However, even if a numeric string as identification information is systematized based on a predetermined rule, in actual operation, the numeric string as identification information is mechanically used as in the check using the check digit as described above. Numbers that cannot be checked are often given.

For example, in an information management system that manages information related to an ID card, a number indicating the number of reissues that is counted up each time an ID card is reissued is often assigned to the last digit or the like. Such reissuance of the ID card is performed an arbitrary number of times at an arbitrary timing. For this reason, it is difficult to check the number indicating the number of ID card reissues in the number string as identification information with a check digit, etc., and the number string as identification information is significantly changed each time the ID card is reissued. It is not preferable in terms of operation. For this reason, in the conventional information management system, the number of digits that are not subject to the check in the numeric string as identification information actually depends on human confirmation without checking the system for erroneous recognition. There is a problem.
JP-A-6-187488

  The present invention solves the above-described problems, and a character recognition device and a character recognition method capable of easily confirming and correcting misrecognition of all the numbers in the number example as a recognition result. The purpose is to provide.

  A character recognition device according to an aspect of the present invention includes a reading unit that reads an image on a recording medium on which a numeric string is described, and a character recognition that recognizes a numeric string from the image on the recording medium read by the reading unit. Means for judging the validity of the numeric value of a predetermined digit in the digit string recognized by the character recognition means based on a preset value, and the judgment means When it is determined that the value is a highly valid value, the first display means for displaying the number string recognized by the character recognition means, and the value of the number of the predetermined digit is low by the determination means. And a second display means for displaying a warning to confirm the number of the predetermined digit in addition to the number string recognized by the character recognition means.

  The character recognition method as one aspect of the present invention is a method of reading an image on a recording medium on which a number string is described, character-recognizing the number string from the read image of the recording medium, and recognizing the number by this character recognition. Judging the correctness of the numeric value of the predetermined digit in the column based on a preset value, and when determining that the numeric value of the predetermined digit is a highly valid value by this determination, A character string that has been character-recognized is displayed on a display device, and if it is determined by the determination that the value of the number of the predetermined digit is not an invalid value, the number of the predetermined digit is confirmed together with the character-recognized number string A warning to that effect is displayed on the display device.

  According to the present invention, it is possible to provide a character recognition device and a character recognition method capable of easily confirming and correcting misrecognition of all the numbers in the number example as the recognition result.

The best mode for carrying out the present invention will be described in detail with reference to the drawings.
FIG. 1 is a diagram schematically showing a configuration example of an information management system 1 according to an embodiment of the present invention.
As shown in FIG. 1, the information management system 1 includes a computer (electronic computer) 11, a reading device 12, a printing device 13, and a server device 14. The information management system 1 manages various types of information associated with identification information as a numerical string organized based on a predetermined rule. In the present embodiment, description will be made assuming an information management system 1 that manages information related to an ID card to which identification information is provided as a number string organized based on a predetermined rule. Further, in the present embodiment, the identification information is given for each ID card or for each individual, and information to be managed in association with the identification information includes, for example, name, address, date of birth, It shall be information such as personal information such as gender.

  The computer 11 is composed of, for example, a general personal computer. The computer 11 recognizes a numeric string as identification information from the image read by the reading device 12, or requests the server apparatus 14 for information corresponding to the identification information using the numeric string as identification information as a key. The printing device 13 performs a printing process based on the information acquired from the server device 14. The computer has interfaces (not shown) for performing data communication with the reading device 12, the printing device 13, and the server device 14, and the reading device 12, the printing device 13, and the server are connected via these interfaces. Data communication with the device 14 is performed.

  The reading device 12 is configured by a general image reading device (scanner) such as a hand scanner, for example. The reading device 12 optically reads an image on a recording medium in which a number string as identification information is described. The reading device 12 optically scans an image of an area including at least an area in which a numeric string as identification information is written on a recording medium, and converts the image into electronic data (image data). The reader 12 outputs the read image data to the computer 11.

The printing device 13 is configured by a printing device such as a general printer, for example. The printing apparatus 13 performs a printing process based on a printing instruction from the computer 11.
The server device 14 is constituted by, for example, a general server computer. The server device 14 is a device that manages information handled by the information management system 1 such as information on an ID card. In response to an inquiry of information using the identification information from the computer 11 as a key, the server device 14 searches the information using the identification information as a key, extracts information corresponding to the identification information, and extracts the extracted information as the above It is to be returned to the computer 11.

FIG. 2 is a block diagram showing a configuration example of the information management system 1 according to the embodiment of the present invention.
As shown in FIG. 1, the computer 11 includes a control unit 21, a storage unit 22, a display unit 23, an operation unit 24, a character recognition (OCR) unit 25, and the like.
The control unit 21 controls the entire computer 11. The control unit 21 includes a CPU (not shown), an internal memory, and the like. The control unit 21 executes various processes based on a control program stored in the storage unit 22 or an internal memory (not shown). That is, the control unit 21 realizes various functions by executing application programs stored in a memory such as the storage unit 22.

  The storage unit 22 stores control data such as a control program for realizing various processes, or processing data. In addition, the storage unit 22 stores, for example, a validity value described later in advance. The display unit 23 displays operation guidance for the operator, a confirmation screen, and the like. The operation unit 24 is used for inputting various operation instructions by an operator. In addition, you may make it comprise the said display part 23 and the said operation part 24 with the display apparatus with a built-in touch panel, for example.

  The character recognition unit 25 recognizes character information such as numbers from image data. In the present embodiment, it is assumed that a function for recognizing characters of each digit constituting a digit string from image data is provided. The character recognition unit 25 is a function realized by the control unit 21 or the like executing an application program stored in a memory such as the storage unit 22.

  The computer 11 includes an interface (not shown) for transmitting / receiving data to / from the reading device 12, an interface (not shown) for transmitting / receiving data to / from the printing device 13, and data to / from the server device 14. And an interface (not shown) for transmitting and receiving data.

  With the configuration as described above, the computer 11 acquires image data by the reading device 12, character-recognizes a numeric string as identification information from the image data acquired from the reading device 12, and as character-recognized identification information The server apparatus 14 is inquired for information based on the numeric string. In the computer 11, information corresponding to a numeric string as identification information acquired from the server device 14 is displayed on the display unit, or printing processing is performed by the printing device 13.

The reading device 12 optically reads an image including at least a numerical string description area from a recording medium on which a numerical string as identification information such as an applicant or an ID card is written, and converts it into image data as electronic data. To do. Further, the reading device 12 supplies the read image data of an area including at least a description area of a numeric string to the computer 11 via an interface (not shown).
The printing device 13 prints application contents including information read from the license C on a predetermined sheet used as an application form (document) based on print data given from the computer 11.

  The server apparatus 14 has a database 41 as shown in FIG. The database 41 stores information relating to individuals, information relating to ID cards, and the like in association with identification information represented by numeric strings. The server device 14 has a control unit (not shown) and an interface (not shown) for performing data communication with the computer 11. With this configuration, the server device 14 extracts information associated with the identification information from the database 41 in response to an inquiry about information using the numeric string as identification information from the computer 11 as a key. The extracted information is returned to the computer 11.

Next, a format example of a numeric string as identification information used in the information management system 1 of the present embodiment will be described.
FIG. 3 is a diagram illustrating a configuration example of a numeric string used as identification information.

  As described above, in the information management system 1 according to the present embodiment, various types of information are stored in the database 41 in association with numeric strings as identification information. For this reason, in order to extract specific information from the database 41, it is necessary to determine a numeric string as identification information. The number string as such identification information can be composed of any number of numbers, but considering the convenience in the operation of the information management system, a format based on a predetermined rule is set in advance. Better to be. Therefore, in the present embodiment, it is assumed that the numeric string as identification information is formed in a systematic format based on a predetermined rule.

  For example, in the example shown in FIG. 3, the numeric string as identification information has a systematized structure according to the following rules (1) to (4).

(1) The identification information is a 12-digit number string.

(2) The first to tenth digits are arbitrary numbers.

(3) The eleventh digit is a number as a check digit of the first to tenth digit strings. The check digit indicates a value (number) calculated based on a predetermined law. In this case, the 11th digit indicates a value (numerical value) calculated from a 1-10 digit string based on a predetermined rule.

(4) The 12th digit is a number that is incremented by one each time a specific process (for example, reissuance of an ID card) is performed, starting from 0.

  In the numerical example organized as shown in FIG. 3, the 1-10 digit number string can be checked based on the number as the 11th digit check digit. However, since the 12th digit number is changed according to a specific process performed at an arbitrary timing such as reissuance of an ID card, it can take an arbitrary value. For this reason, in a systematic number string as shown in FIG. 3, digits (twelfth digit) that cannot be checked by the check digit cannot be checked only by the number string itself.

  In the present embodiment, in the numeric string organized as described above, digits of digits that cannot be checked by the check digit (digits that cannot be checked from the numeric sequence itself) are checked based on preset conditions. As a condition for such a check, for example, a validity value can be set for a digit of a digit whose validity cannot be checked from the numeric string itself.

  That is, the information management system according to the present embodiment checks a specific digit in the numeric string as a recognition result based on the validity value as described above, and determines whether or not the digit is likely. To do. Further, the validity value may designate a numerical range for a number, may designate a specific number, or designates a condition that can be logically determined by a computer. It may be a thing. The validity value as described above is data set (stored) in advance in the storage unit 22 of the computer 11 or an internal memory (not shown) in the control unit 21.

  For example, as described above, it is assumed that the 12th digit is changed according to the reissue of the ID card. In this case, the number of ID card reissues is usually “0” (no reissue), followed by “1” (one reissue number) and “2” (two reissue times). ) In order. Furthermore, although it depends on the operation status of the ID card, “3” or more as the number of ID card reissues is rare. In other words, since the number of the 12th digit as the number of reissues of the ID card is rarely “3” or more, if the number of the 12th digit is “3” or more as a recognition result, The 12th digit is likely to be misrecognized (that is, less legitimate).

In such a case, “3” is set as the validity value, it is determined whether or not the 12th digit is “3” or more, and the 12th digit is “3” or more. In this case, it is determined that there is a high possibility that the 12th digit is erroneously recognized (low legitimacy). As a result of this determination, it is possible to warn the operator about the recognition result determined to have a high possibility of misrecognition, and to allow the operator to confirm the numeric string as the recognition result. As a result, the operator can confirm by paying attention to a number string having a high possibility of erroneous recognition or a number having a high possibility of erroneous recognition in the number string. As a result, it is possible to improve the efficiency of the identification information input process or the entire process in the information management system.
In this way, by setting a validity value for determining whether or not a certain digit number is probable according to the probability of appearing as a valid number and the probability of appearing as a misrecognition, the recognition result is It is possible to easily check whether the possibility of erroneous recognition is high.

  Moreover, as a tendency of misrecognition in the character recognition process, for example, there is a tendency that “0” is misread as “8”. This is a misrecognition that easily occurs when the recording medium on which “0” is written is dirty. The correctness value may be set in consideration of the tendency of erroneous recognition in such character recognition processing. For example, as described above, when the number indicating the ID card reissue number is the 12th digit, the 12th digit is most likely “0”, and the 12th digit is “8”. "Is extremely low. In such a case, by setting “3” as the validity value, when the 12th digit is recognized as “8”, the 12th digit is greater than or equal to the validity value “3”. Therefore, it is possible to determine that there is a high possibility of misreading.

  In this way, the validity for determining whether or not a certain digit number is probable according to the probability of appearing as a valid number, the probability of appearing as a misrecognition, and the tendency of misrecognition in character recognition processing. By setting a value, it is possible to easily check a recognition result having a high possibility of erroneous recognition according to a tendency of erroneous recognition in the character recognition process.

Next, identification information recognition processing (input processing) in the information management system 1 will be described.
FIG. 4 is a flowchart for explaining the flow of identification information recognition processing (input processing) in the information management system 1. 5, FIG. 6, FIG. 7 and FIG. 8 are diagrams showing display examples of the display unit 23 in the identification information recognition processing stage. Here, the display unit 23 and the operation unit 24 of the computer 11 are configured by a display device with a built-in touch panel, and an operation instruction from the operator touches a touch key displayed on the display device. It shall be input by.

  First, when the power of the computer 11 is turned on, the control unit 21 of the computer 11 executes a startup process and performs a process of confirming a connection state with the reading device 12, the printing device 13, and the server device 14. . When the connection with the reading device 12, the printing device 13, and the server device 14 is established and the computer 11 is started, the control unit 21 displays an acceptance screen for receiving identification information input processing on the display unit 23 as an initial screen. Display (step S1). For example, FIG. 5 shows a display example of the reception screen. In the example of the reception screen illustrated in FIG. 5, an operation guide for reading an image of an area in which a number string as identification information is read by the reading device 12 is displayed. In the reception state in which such a reception screen is displayed, the information management system 1 can start reading an image on a recording medium in which identification information is described.

  In the reception state as described above, the operator reads an image of an area including an area in which a numeric string as identification information on the recording medium is described as image data by the reading device 12 (step S2). Thus, the image data of the area including the area where the read digit string is described is transmitted from the reading device 12 to the computer 11.

  The computer 11 that has received the image data from the reading device 12 performs character recognition processing for recognizing a numeric string from the image data received from the reading device 12 by the character recognition unit 25 (step S3). When the character string is recognized by the character recognition processing by the character recognition unit 25, the control unit 21 checks the number string obtained as a recognition result with a check digit set to a predetermined digit (step S4). .

  For example, when a numeric string as shown in FIG. 3 is used as identification information, the control unit 21 checks the 1st to 10th digits based on the 11th digit. That is, for the systematic number string as shown in FIG. 3, the control unit 21 calculates the first to tenth digits based on a predetermined rule, and the calculation result is the eleventh digit number. It is determined whether or not they match, and if they match, it is determined that the recognition result is up to 11 digits of the numeric string, and if they do not match, it is determined that there is a misrecognition up to 11 digits of the numeric string.

  When it is determined that the recognition result is incorrect as a result of the check by the check digit as described above, that is, when erroneous recognition is detected by the check by the check digit, the control unit 21 performs the step S2 or the step S3. Returning to step S5, the number string recognition process is performed again (step S5). In addition, when erroneous recognition in the numeric string as the recognition result is detected by the check by the check digit as described above, the control unit 21 may end the process with the recognition process of the numeric string as an error. .

  Further, when it is determined that the number string as the recognition result is correct as a result of the check by the check digit as described above, that is, when the erroneous recognition is not detected by the check by the check digit, the control unit 21 further Processing for checking the validity of the digits of digits that cannot be checked with the check digit (validity check) is performed (step S6). In this validity check, the control unit 21 reads the validity value stored in the storage unit 22 and the like, and the digit number that is the target of the validity check seems to be correct based on the read validity value. It is determined whether or not (the legitimacy is high).

  For example, when a numeric string as shown in FIG. 3 is used as the identification information and the validity value is set to “3”, the control unit 21 performs the 12th digit of the numeric string as the recognition result. Check whether the number is less than “3” or greater than “3”. That is, when the validity value is set to “3” for the systematic number string as shown in FIG. 3, the control unit 21 determines that the 12th digit number is less than the validity value “3”. Or if the 12th digit is less than the validity value “3”, the 12th digit seems to be correct (highly valid). If the 12th digit number is greater than or equal to the validity value “3”, it is determined that the 12th digit number is not correct (the validity is low, that is, the possibility of erroneous recognition is high).

  When the validity check as described above is successful, that is, when it is determined by the validity check that the digit of the digit to be checked is correct, the control unit 21 provides the operator with a number string as the recognized identification information. The confirmation screen for confirmation is displayed by the display unit 23 (step S7). At this time, the control unit 21 also displays on the display unit 23 a confirmation key for confirming a numeric string as a recognition result to the operator.

  For example, FIG. 6 shows a display example of a confirmation screen displayed by the display unit 23 when the validity check is successful. In the example of the confirmation screen shown in FIG. 6, a display area in which a number string as recognized identification information is displayed, a confirmation key for confirming a number string as a displayed recognition result, and a number as a displayed recognition result The cancel key for canceling the column is displayed.

  When the validity check as described above fails, that is, when it is determined by the validity check that the number of digits to be checked is likely to be erroneously recognized, the control unit 21 recognizes the recognized identification information. Is displayed on the display unit 23 as a warning screen to confirm the digits of the digit subjected to the validity check (step S8). At this time, the control unit 21 also displays on the warning screen a number key for correcting and inputting the digit of the digit subjected to the validity check and the guidance indicating the digit subjected to the validity check on the warning screen. (Step S9).

  For example, FIG. 7 shows a display example of a warning screen displayed by the display unit 23 when the validity check fails. In the example of the warning screen shown in FIG. 7, in addition to the display content of the confirmation screen shown in FIG. 6, each numeric key from “0” to “9” and a number determined to have a high possibility of erroneous recognition (valid An arrow indicating a digit subjected to the sex check) and a guide to correct if the number of the arrow is incorrect are displayed. In a state in which such a warning screen is displayed, the operator checks the number indicated by the arrow, and corrects the correct number using the number keys if there is a correction.

  When correcting the number on the warning screen as described above, the operator selects and inputs the correct numeric key from the numeric keys displayed on the warning screen. When correction input is performed by the operator, the control unit 21 detects the number input by the operator and replaces the target number with the number input by the operator (step S10).

  For example, FIG. 8 shows a display example when the numeric key “0” is input in the display state shown in FIG. When the operator inputs the numeric key “0” in the warning screen display state as shown in FIG. 7, the control unit 21 displays the warning screen displayed by the display unit 23 in the display state as shown in FIG. And That is, the digit of the correction target displayed as “8” on the warning screen shown in FIG. 7 is replaced with “0” when the operator inputs the numeric key “0”. Is displayed as shown in FIG.

  Further, when the operator inputs a confirmation key on the display screen as shown in FIG. 6, FIG. 7, or FIG. 8, the control unit 21 confirms the number string as the recognition result displayed on the screen (step S11). When the number string as the recognition result is confirmed, the control unit 21 of the computer 11 transmits a signal requesting information corresponding to the confirmed number string to the server device 14. Upon receiving the request for information corresponding to such a numeric string, the server device 14 extracts information corresponding to the numeric string from the database 41 and transmits it to the computer 11. Thereby, in the said computer 11, the information corresponding to the number sequence as a recognition result of the number sequence described in the recording medium can be acquired.

  As described above, in the present embodiment, an image on a recording medium is read by a reading device, a number sequence described on the recording medium is recognized from the image read by the reading device, and a predetermined number sequence in the recognized number sequence is determined. It is determined whether or not a digit number is likely to be erroneously recognized based on a preset validity value, and this determination determines that the predetermined digit number is likely to be erroneously recognized. In such a case, a warning is displayed to confirm the predetermined digit number together with the recognized number string. As a result, even if the digit number cannot be checked for erroneous recognition based on the configuration of the numeric string, the operator can be warned if there is a high possibility of erroneous recognition.

  Further, when it is determined that there is a high possibility that the predetermined digit number is erroneously recognized, a warning is displayed to confirm the predetermined digit number together with the recognized digit string, and the correction to the predetermined digit number is performed. When the number is input, the predetermined number displayed is replaced with a correction number and displayed. As a result, it is possible to make the operator confirm the number that is highly likely to be erroneous recognition and the corrected number.

  In addition, when it is determined that the predetermined digit number is highly likely to be erroneously recognized, a correction number corresponding to the predetermined digit number is displayed together with a warning that the predetermined digit number is confirmed together with the recognized digit string. The numeric key for designating the number is displayed, and the predetermined digit is replaced with the corrected number in the displayed numeric string in response to the input of the corrected number for the predetermined digit. . Thereby, it is possible to easily perform correction input for a number that is highly likely to be erroneous recognition.

The figure which shows schematically the structural example of the information management system which concerns on embodiment of this invention. The block diagram which shows the structural example of an information management system. The figure which shows the structural example of the numerical sequence as identification information. The flowchart for demonstrating the input process of the identification information in an information management system. The figure which shows the example of a display of the display part in an initial state (acceptance state). The figure which shows the example of a display when the correctness of the recognized number string is confirmed. The figure which shows the example of a display when the correctness with respect to the number of the specific digit in the recognized number sequence cannot be confirmed. The figure which shows the example of a display when correct | amending the digit of the digit which cannot confirm the correctness in the recognized number string.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Information management system, 11 ... Computer (electronic computer), 12 ... Reading apparatus, 13 ... Printing apparatus, 14 ... Server apparatus, 21 ... Control part, 22 ... Memory | storage part, 23 ... Display part, 24 ... Operation part, 25 ... Character recognition part (OCR processing part), 41 ... Database

Claims (6)

Reading means for reading an image on a recording medium on which a numerical string is described;
Character recognition means for recognizing a character string from an image of the recording medium read by the reading means;
A determination unit that determines the validity of a predetermined digit number value in the digit string recognized by the character recognition unit based on a preset value;
A first display unit for displaying a digit string recognized by the character recognition unit when the determination unit determines that the numeric value of the predetermined digit is a highly valid value;
When it is determined by the determination means that the numeric value of the predetermined digit is a low-validity value, a warning that the number of the predetermined digit is confirmed in addition to the numeric string recognized by the character recognition means Second display means for displaying;
A character recognition device comprising: Further, when a correction number for the predetermined digit is designated in response to the warning displayed by the second display unit, the predetermined digit of the number string displayed by the second display unit is specified. A third display means for displaying a number by replacing the number with a correction number;
The character recognition apparatus according to claim 1, wherein: The second display means displays a number key for designating a correction number for the predetermined digit number, together with a warning to confirm the predetermined digit number,
In the number string displayed by the second display means, the third display means, when a correction number for the predetermined number is designated by the numeric key displayed by the second display means, The predetermined number is replaced with a correction number and displayed.
The character recognition device according to claim 2, wherein: Read the image on the recording medium on which the numeric string is written,
Recognize the character string from the read image of the recording medium,
Judging the validity of the numeric value of a predetermined digit in the numeric string recognized by the character recognition based on a preset value,
When it is determined by this determination that the value of the number of the predetermined digit is a highly valid value, the character recognized number string is displayed on a display device,
When it is determined by the determination that the value of the number of the predetermined digit is not a low-validity value, a warning that the number of the predetermined digit is confirmed together with the character-recognized number string is displayed on the display device.
A character recognition method characterized by the above. Further, when a correction number for the predetermined digit number is designated in accordance with the warning displayed on the display device, the correction for which the predetermined digit number is designated in the digit string displayed on the display device Display with the number of
The character recognition method according to claim 4, wherein: In addition, with a warning that the number of the predetermined digit is confirmed, a number key for displaying a correction number for the number of the predetermined digit is displayed,
When a correction number for the predetermined digit is designated by the number key displayed on the display device, the predetermined digit is replaced with a correction number in the number string displayed by the display device. To
The character recognition method according to claim 5, wherein:

Images