JP6255929B2 - Information processing apparatus, information processing apparatus control method, and program - Google Patents

Description

The present invention relates to an information processing apparatus, a control method for the information processing apparatus, and a program, and in particular, an error in characters when reading characters from image data generated due to characters entered by a user overlapping an entry frame. the recognition relates capable of improving the accuracy of correcting mechanism.

  Conventionally, there is a character recognition technique that converts an imaged character into a text code. In particular, there is also a technology for recognizing handwritten characters on an application form or a form with an entry frame.

  In addition, there is a technology that not only recognizes the frame and recognizes the character inside the frame part, but also performs character recognition (separates the character and the frame) by deleting the frame.

  Patent Document 1 proposes a method for separating a character and a frame using the brightness of a multi-value image in order to separate the character from the frame or the background color.

  Japanese Patent Application Laid-Open No. H10-228867 proposes a method of judging a structure (separating a character area) by labeling pixels judged to have the same density, paying attention to a density difference between a character area and a non-character area.

Japanese Patent Application Laid-Open No. 2001-076096 Japanese Patent Application Laid-Open No. 09-016713

  However, when the entered characters are in contact with the entry frame, the characters entered with a writing instrument such as a ball pennage are also shaded on the entered characters due to the writing pressure, the ink amount, etc. Even if technology is used, it is difficult to accurately separate characters and frames, and even if they can be separated, there are cases in which it is difficult to create a binary image at a level that satisfies the recognition performance of OCR (Optical Character Recognition). It was.

This onset Ming, provides a misrecognition of characters when reading a character from the image data characters entered by the user is caused to overlap the input frames, a capable of improving the accuracy of correcting mechanism It shall be the object of the present invention to.

The present invention is filled, a predetermined character, and management means for managing in association with the other characters, the document medium with mark frames to fill the character from the data of the image data, by a user to the entry box determining a reading means for reading a character which is a pre-SL read characters read by means of a predetermined character, and, if the predetermined character read are overlapped in the entry box and place The determination means and the determination means determine that the character read by the reading means is the predetermined character and that the read predetermined character overlaps the entry frame at a predetermined position. and in response to, and determination means for said predetermined character read by said reading means, determines that replacing the predetermined in association with the character the other which is managed by said management means characters Characterized in that it obtain.

Further, the present invention includes a Jo Tokoro character, a method of controlling an information processing apparatus including a management unit to manage in association with other characters, reading means of the information processing apparatus, to fill the character a document medium example Bei mark frames from the data of the image data, the reading process reads the characters entered by the user in the entry boxes, determining means of the information processing apparatus, the character read by said reading step predetermined of a character, and a not Luke determining step of determining overlap in the said predetermined character is the entry frame and a predetermined position read, the determining unit of the information processing apparatus, in the determination step, In response to determining that the character read in the reading step is the predetermined character and that the read predetermined character overlaps the entry frame at a predetermined position. Read in Said predetermined character, characterized in that it comprises a determination step of replacing a decision to the other characters that are more manageable before Symbol management means in association with the predetermined character.

The present invention includes a predetermined character, a read program executable by an information processing apparatus including a management unit to manage in association with the other characters, the information processing apparatus, an entry box to enter text Reading means for reading characters entered by the user in the entry frame from image data obtained by converting the provided document medium into data, and the characters read by the reading means are predetermined characters, and the read determination means for determining a predetermined character Luke not overlap in the entry box and a predetermined position, by the determination unit, a character read is predetermined by the reading means, and the read and in response to said predetermined character is determined to overlap in the entry box and place, the predetermined character read by said reading means, before in association with the predetermined character Characterized in that to function as a determining means for replacing a decision to the other characters that are more manageable to the management unit.

According to the present invention, erroneous recognition of a character when reading a character from the image data character entered by the user is caused to overlap the input frames, a capable of improving the accuracy of correcting mechanism Can be provided.

It is a figure which shows an example of the system configuration | structure concerning embodiment of this invention. It is a block diagram which shows an example of the hardware constitutions of the apparatus concerning embodiment of this invention. It is an example of the main flowchart which shows the whole flow which concerns on the 1st Embodiment of this invention. It is an example which shows the data structure of the OCR correction information table concerning embodiment of this invention. It is an example of the flowchart of the presence determination of the contact character performed with the client terminal 101 which concerns on the 1st Embodiment of this invention. It is an example which shows the data structure of the entry frame information table concerning embodiment of this invention. It is an example which shows the binary image of the entry frame and entry character concerning embodiment of this invention. It is an example which shows the binary image at the time of carrying out the noise removal process of the entry frame concerning embodiment of this invention. It is an example which shows the binary image of the side removed when the entry frame concerning embodiment of this invention performed the noise removal process. It is an example which shows the data structure of the contact information table concerning embodiment of this invention. It is an example of the flowchart of the gray frame deletion process performed with the client terminal 101 which concerns on the 1st Embodiment of this invention. It is an example which shows each gray value frequency distribution of the entry frame and the entry character concerning embodiment of this invention. It is an example which shows the image at the time of binarization processing which erases the entry frame concerning embodiment of this invention, and leaves only an entry character. It is an example of the flowchart of the frame contact part determination process performed with the client terminal 101 which concerns on the 1st Embodiment of this invention. It is an example of the flowchart of the OCR correction determination process performed with the client terminal 101 which concerns on the 1st and 2nd embodiment of this invention. It is an example of the flowchart of the frame contact part determination process performed with the client terminal 101 which concerns on the 2nd Embodiment of this invention. It is an example of the main flowchart which shows the whole flow which concerns on the 2nd Embodiment of this invention. It is a figure which shows an example of the acquisition method of the rectangle of a character when the entry character is contacting the entry frame in step S301 of FIG. 3, and when not contacting the entry frame.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
<Description of FIG. 1>

  FIG. 1 is a diagram illustrating an example of a system configuration according to an embodiment of the present invention.

The client terminal 101 (information processing apparatus) of the present invention is, for example, a personal computer, has a network communication function, and can communicate (connect) with the server 102 via the network 1 (LAN, WAN, etc.). .
<Description of FIG. 2>

  FIG. 2 is a block diagram illustrating a hardware configuration example of the client terminal 101 and the server 102.

  As shown in FIG. 2, the client terminal 101 server 102 includes a CPU (Central Processing Unit) 201, a RAM (Random Access Memory) 203, a ROM (Read Only Memory) 202, an input controller 205, and a video controller via a system bus 204. 206, a memory controller 207, a communication I / F controller 208, and the like are connected.

  The CPU 201 comprehensively controls each device and controller connected to the system bus 204.

  Further, the ROM 202 or the external memory 211 will be described later, which is necessary for realizing the functions executed by each server or each PC, such as BIOS (Basic Input / Output System) and OS (Operating System) which are control programs of the CPU 201. Various programs are stored. Further, information necessary for carrying out the present invention is stored. The external memory may be a database.

  The RAM 203 functions as a main memory, work area, and the like for the CPU 201. The CPU 201 implements various operations by loading a program or the like necessary for executing the processing from the ROM 202 or the external memory 211 to the RAM 203 and executing the loaded program.

  The input controller 205 controls input from a keyboard (KB) 209 or a pointing device such as a mouse (not shown).

  The video controller 206 controls display on a display device such as the display 210. The display device may be a display device such as a liquid crystal display. These are used by the administrator as needed.

  The memory controller 207 is an external storage device (hard disk (HD)), flexible disk (FD), or PCMCIA (Personal Computer) that stores a boot program, various applications, font data, user files, editing files, various data, and the like. Controls access to an external memory 211 such as a Compact Flash (registered trademark) memory connected to a Memory Card International Association (Card Memory) card slot via an adapter.

  The communication I / F controller 208 is connected to and communicates with an external device via the network 104, and executes communication control processing in the network. For example, communication using TCP / IP (Transmission Control Protocol / Internet Protocol) is possible.

  The CPU 201 can display on the display 210 by executing an outline font rasterization process on a display information area in the RAM 203, for example. Further, the CPU 201 enables a user instruction using a mouse cursor (not shown) on the display 210.

Various programs to be described later for realizing the present invention are recorded in the external memory 211 and executed by the CPU 201 by being loaded into the RAM 203 as necessary. Furthermore, definition files and various information tables used when executing the program are also stored in the external memory 211, and a detailed description thereof will be described later.
<Description of First Embodiment>
<Description of FIG. 3>

  FIG. 3 is an example of a main flowchart showing an overall flow of the OCR process executed by the client terminal 101 according to the first embodiment of the present invention. Each step of the flowchart of FIG. 3 is executed by the CPU 201 of the client terminal 101. In this embodiment, electronic data (also referred to as image data) obtained by reading a document medium (application form, form, etc.) handwritten by the user in the entry frame with a scanner or the like is stored in the external memory 211 of the client terminal 101 in advance. (Image data obtained by converting a document medium in which characters are entered by a user into data in an entry box for entering characters is stored), and the processing in FIG. 3 is performed by OCR processing the electronic data.

  In step S301, normal OCR processing is performed. In the present embodiment, it is assumed that the handwritten numbers entered in the entry frame are OCRed. OCR processing by a program is executed by the CPU 201, and the OCR result of a number entered by handwriting is stored in the RAM 203. In this embodiment, the OCR result 1003 in the contact information table in FIG. 10 is stored.

Note that step S301 is an example of reading means for reading characters entered by the user from image data obtained by converting a document medium in which characters are entered by a user into a form having an entry frame for entering characters in the present invention. It is.
<Description of FIG. 10>

  FIG. 10 shows an example of the data structure of the contact information table according to the embodiment of the present invention. This information table is an information table for storing how much and how much the characters entered in the entry frame (rectangular frame) are in contact with the frame (the expression “contact” in this embodiment is also referred to as “overlap”). . 4, 6, and 10 may be stored in the external memory 211 of the client terminal 101. In this case, the external memory 211 is an example of a storage unit that stores information on characters to be replaced with other characters when the characters overlap the rectangular frame at a predetermined position in the present invention. In the case of storing in the RAM 203, the RAM 203 is an example of a storage unit.

  Specifically, it consists of a frame number 1001, the number of digits 1002, an OCR result 1003, contact presence / absence 1004, a contact location 1005, and a contact ratio 1006. In the present embodiment, the contact portion 1005 is the upper, lower, left, and right sides of the entry frame. However, the contact portion 1005 may be subdivided such as an upper left portion and a lower left portion. In the present embodiment, the contact ratio 1006 is the ratio of black text pixels to the frame. However, the contact ratio 1006 may be scored by weighting according to the contact location.

  Returning to the description of FIG.

In step S302, it is determined whether or not the designated character is included in the OCR result. For the determination, an OCR correction information table shown in FIG. 4 is used. If there is a character (“1”, “5”, “7” in the example of FIG. 4) described in the OCR result character (401) of this information table, the process proceeds to step S303. If it does not exist, the process ends.
<Description of FIG. 4>

  FIG. 4 shows an example of the data structure of the OCR correction information table according to the embodiment of the present invention. This information table shows information on how much the handwritten text is in contact with which part of the frame, and what character the OCR result is likely to be erroneously recognized as (if the character overlaps the rectangular frame at a predetermined position, This is an information table for storing character information to be replaced with characters.

  Specifically, OCR result character 401, contact location (contact location between entry frame and entry character) 402, contact ratio (overlapping state determined to be replaced as character information) 403, recognition correction character (misrecognized) Character to be converted in this case) 404.

  In the present embodiment, the contact location 402 is the top, bottom, left, and right of the entry frame, but it may be subdivided such as an upper left side and a lower left side. In the present embodiment, the contact ratio 403 is the ratio of black text pixels to the frame. However, the contact ratio 403 may be scored by weighting according to the contact location.

  This information table may be manually created in advance by the user in advance, or may be automatically created by collecting information in the process of performing OCR separately and correcting erroneous recognition artificially.

  Returning to the description of FIG.

  In step S303, a process for determining the presence or absence of a contact character is performed. Details of this processing will be described with reference to the flowchart of FIG.

  In step S304, a process for determining the presence or absence of a contact character is performed. In this processing, based on the information 1004 in the contact information table shown in FIG. 10, it is determined whether a contact location exists in the character of the frame. If there is a touching part, the process proceeds to step S305. If it does not exist, the process ends.

  In step S305, a process is performed for determining whether or not the OCR result of the portion designated as the contact character portion in the contact information table shown in FIG. 10 matches the correction target character specified in the OCR correction information table shown in FIG. Do. If there is a matching part, the process proceeds to step S306. If it does not exist, the process ends.

  In step S306, a gray frame erasing process is performed. Details of this processing will be described with reference to the flowchart of FIG.

  In step S307, a frame contact portion determination process is performed. Details of this processing will be described with reference to the flowchart of FIG.

In step S308, an OCR correction determination process is performed. Details of this processing will be described with reference to the flowchart of FIG.
<Description of FIG. 5>

  FIG. 5 is an example of a flowchart of the presence / absence determination of a contact character executed by the client terminal 101 according to the first embodiment of the present invention. Each step of the flowchart of FIG. 5 is executed by the CPU 201 of the client terminal 101.

In step S501, frame size information acquisition processing is performed. Specifically, the information in the frame information table shown in FIG. 6 is acquired and stored in the RAM 203.
<Description of FIG. 6>

  FIG. 6 shows an example of the data structure of the entry frame information table according to the embodiment of the present invention. This information table is an information table for storing the position information of the entry frame on the image for each entry frame.

Specifically, it consists of a frame number 601, a digit number 602, a frame coordinate 603, a frame width (frame line width) 604, and a frame erase threshold 605.
“Up” and “Down” of the frame coordinates 603 represent the y coordinates of the upper surface (line) and the lower surface (line) of the entire frame number 601, and “left” and “right” of the frame coordinates 603 are , The x coordinate of the leftmost surface (line) and the rightmost surface (line) of the entire frame number 601 is represented.

  In this embodiment, the frame coordinates 603 and the frame line width 604 are the upper, lower, left, and right sides of the entry frame (the entire frame number 601). However, the information is stored for each frame (the coordinates of the entry frames are stored). It does not matter.

  If the size of the entry frame is determined, it can be determined and given in advance.

  Returning to the description of FIG.

  In step S502, a noise removal parameter calculation process is performed based on the information in the frame information table shown in FIG. 6 obtained in step S501. Specifically, the size corresponding to the inscribed rectangle of each entry frame (the range of the rectangle not inscribed in the entry frame) is obtained by calculation. In addition, when the size of the entry frame is determined, it can be determined and given in advance. In addition, the noise here is a handwritten character entered without touching the entry frame.

  In step S503, noise removal processing is performed based on the noise removal parameter obtained in step S502. By this processing, handwritten characters entered without touching the entry frame are removed as noise, and only the characters in contact with the frame remain in the image.

  As a specific example, when the noise removal processing is executed, an image as shown in FIG. 8 is created from an image in which characters are entered in an 8-digit entry box shown at 701 in FIG. At the same time, the image shown in FIG. 9 is also generated as the image on the side deleted as noise and stored in the RAM 203. In this embodiment, the noise removal process is executed in 701 in FIG. 7, but this process is actually executed for entry boxes other than 701.

  In step S504, OCR processing is performed on the image after noise removal. The processing result is temporarily stored in the RAM 203.

  In step S505, the result of the OCR process in step S504 is referred to and a process for determining whether the OCR result is a space character is performed. Step S505 replaces the character read by the reading unit based on the character information stored in the storage unit and the overlapping state of the character entry box entered by the user in the present invention. It is an example of the determination means which determines whether or not.

  When there is a recognition result other than a space character in the OCR result 1003 of the contact information table, it becomes a space character, and it is determined that a character that does not touch the entry frame is entered and noise is removed. In the processes in steps S505 and S506, the process is repeated for the number of digits (602 in FIG. 6) from the first character of the frame. If it becomes a space character, the process proceeds to step S507. Otherwise, the process proceeds to step S506.

  In step S506, a contact presence flag is set in the contact presence / absence 1004 of the contact information table in FIG. At this time, a flag may be set unconditionally, or a flag may be set on the condition that the recognition result in step S504 matches the OCR result 1003 of the contact information table in FIG.

  In step S507, processing for determining whether there is a remaining frame of the entry frame is performed. If there is still an entry frame, the process proceeds to step S505. If it does not exist, the contact character presence determination process is terminated.

  FIG. 7 is an example showing a binary image of an entry frame and entry characters according to the embodiment of the present invention. Further, 701 corresponds to the frame number (601, 1001) 1 of the information table in FIGS. 6 and 10 in this embodiment, and 702 corresponds to the number of digits of 602 in FIG. 6 and 1002 in FIG. In the description of FIG. 8 and FIG. 9 of the present embodiment, description will be made by taking the frame number 701 as an example.

  FIG. 8 is an example showing a binary image when the entry frame according to the embodiment of the present invention is subjected to noise removal processing. This corresponds to an example of the execution result of step S503 in FIG.

FIG. 9 shows an example of a binary image on the removed side when the entry frame according to the embodiment of the present invention is subjected to noise removal processing. This corresponds to an example of the execution result of step S503 in FIG. The image shown in FIG. 9 has a feature that only the entered characters remain.
<Description of FIG. 11>

  FIG. 11 is an example of a flowchart of a gray frame erasing process executed by the client terminal 101 according to the first embodiment of the present invention. Each step of the flowchart of FIG. 11 is executed by the CPU 201 of the client terminal 101.

  In step S1101, a gray value distribution acquisition process of the noise-removed image is performed. Specifically, for the portion of the binary image shown in FIG. 8 where black pixels are present, a process of acquiring the gray value from the gray image before binarization is performed. The processing result is temporarily stored in the RAM 203.

  Finally, data corresponding to the upper graph of the gray value frequency graph shown in FIG. 12A is acquired. As a result, the gray value frequency distribution of the entry frame can be acquired although the gray value of the partially touched entry character remains.

  In step S1102, gray value distribution acquisition processing is performed on the noise-removed image.

  Specifically, for the portion of the binary image shown in FIG. 9 where black pixels exist, processing for acquiring the gray value from the gray image before binarization is performed. The processing result is temporarily stored in the RAM 203.

Finally, data corresponding to the lower graph of the gray value frequency graph shown in FIG. 12B is acquired. Thereby, the gray value frequency distribution of the entered characters can be acquired.
In the processing of step S1101 and step S1102, in the present invention, from the information read by the reading means, the density of the character determined not to overlap the entry frame and the character determined to overlap the entry frame are determined. It is an example of the acquisition means which acquires the density | concentration of the said entry frame containing a character, respectively.

  In step S1103, the binarization threshold value used in the subsequent step S1104 is calculated. The frequency peak value of the gray value distribution obtained in step S1101 and step S1102 is adopted as a threshold value. In step S1103, based on the density difference between the density of the character determined not to overlap the entry frame acquired by the acquisition unit and the density of the entry frame in which the character is entered in the present invention, the image data Is a threshold value determining unit that determines a threshold value for binarizing and setting only the character to a black pixel.

  Specifically, the intermediate value (1203) of both peaks (1201 and 1202), the lowest frequency between the two frequency peaks in the graph of FIG. 12a, or the total frequency of FIG. The method for determining the threshold, such as adopting a value, can be changed for each target sample.

  In step S1104, a frame erase image creation process is performed. More specifically, the OCR target gray image is binarized according to the threshold value (eg, 1203) determined in step S1103. This makes it possible to obtain an image in which the entry frame disappears and the handwritten characters remain. FIG. 13 is an example of an image after this processing is performed. Note that step S1104 is an example of binarization means that binarizes the image data using the threshold value determined by the threshold value determination means so that only the characters are black pixels.

  FIG. 12 is an example showing the gray value frequency distribution of the entry frame and the entry character according to the embodiment of the present invention.

  In this embodiment, the graph is expressed as a graph. However, for example, the distribution can be stored as a function so that the processing in step S1103 in FIG. 11 is facilitated.

  FIG. 13 is an example showing an image when binarization processing is performed so that the entry frame according to the embodiment of the present invention is erased and only the entered characters are left. This corresponds to an example of the execution result of step S1104 in FIG.

  By executing the processing of FIG. 11, it becomes easy to determine at what position and at what rate the entered character and the entry frame are in contact with each other in the processing of FIG.

In this embodiment, the processing of FIG. 14 is executed after the processing of FIG. 11. However, as another embodiment, the image (for example, FIG. 13) obtained by the processing of FIG. Character information acquired by reading characters entered by the user from the image data binarized by the binarization means may be used as the OCR result.
<Description of FIG. 14>

  FIG. 14 is an example of a flowchart of a frame contact portion determination process executed by the client terminal 101 according to the first embodiment of the present invention. Each step of the flowchart of FIG. 14 is executed by the CPU 201 of the client terminal 101.

  In step S1401, a contact location specifying process using a frame erased image is performed for each digit of the entry frame.

  Specifically, based on the frame erased image as shown in FIG. 13 and the information in the entry frame information table in FIG. 6, it is calculated where the black pixel of the entry character exists on the entry frame coordinates. The calculation result is stored in the contact location 1005 of the contact information table in FIG.

  In step S1402, a contact status specifying process using a frame erased image is performed for each digit of the entry frame. Specifically, based on the frame erased image as shown in FIG. 13 and the information (603 and 604) in the FIG. 6 entry frame information table, how many black characters of entry characters exist on the entry frame coordinates. Calculate The calculation result is stored in the contact ratio 1006 of the contact information table in FIG.

In step S1403, a process is performed to determine whether there is a digit that has not yet been subjected to the contact portion determination process in the entry frame. If there is a digit for which the contact portion determination processing has not yet been performed, the process proceeds to step S1401. If it does not exist, the contact portion determination process ends.
<Description of FIG. 15>

  FIG. 15 is an example of a flowchart of an OCR correction determination process executed by the client terminal 101 according to the first and second embodiments of the present invention. Each step of the flowchart of FIG. 15 is executed by the CPU 201.

  In step S1501, a determination process is performed as to whether or not the OCR correction condition matches for each digit of the entry box.

  Specifically, with respect to each digit (1002) of the frame contact information table of FIG. 10, the OCR result 1003 and the OCR result character 401 match (predetermined character) and the presence / absence of contact 1004 is compared with FIG. Yes, it is determined whether the contact location 1005 matches the contact location 402 (contacts the entry frame at a predetermined position), and the contact rate 1006 exceeds the contact rate 403 (overlaps a predetermined rate). If the determination results match, the process proceeds to step S1502. If not, the process proceeds to step S1503.

  In step S1502, the OCR result replacement process executed in step S301 in FIG. 3 is performed for each digit of the entry box.

  Step S1502 is a replacement unit that replaces the character read by the reading unit with another character stored in the storage unit when it is determined that the character should be determined to be replaced by the determination unit in the present invention. It is an example.

  Specifically, a process of replacing the OCR result of the corresponding part with the character described in the recognized corrected character 404 of the OCR correction information table in FIG. 4 is performed.

  In step S1503, a process is performed to determine whether there are any digits that have not yet been subjected to the OCR correction determination process in the entry frame. If there is a digit for which OCR correction determination processing has not yet been performed, the process proceeds to step S1501. If not, the OCR correction determination process is terminated.

In the above-described embodiment, a gray image is used together with a binary image. However, for example, by performing a flowchart of FIG.
<Description of Second Embodiment>
<Description of FIG. 17>

  FIG. 17 is an example of a main flowchart showing an overall flow of the OCR process executed by the client terminal 101 according to the second embodiment of the present invention. Each step of the flowchart of FIG. 17 is executed by the CPU 201 of the client terminal 101. In this embodiment, electronic data obtained by reading a document medium (application form, form, or the like) handwritten by the user in the entry frame with a scanner or the like is stored in advance in the external memory 211 of the client terminal 101. FIG. In this process, the electronic data is subjected to OCR processing. Note that the same processing as in FIG. 3 is assigned the same step number as in FIG.

  In step S301, normal OCR processing is performed.

  In step S302, it is determined whether or not the designated character is included in the OCR result. For the determination, an OCR correction information table shown in FIG. 4 is used. If there is a character (“1”, “5”, “7” in the example of FIG. 4) described in the OCR result character (401) of this information table, the process proceeds to step S1701. If it does not exist, the process ends.

  In step S1701, frame contact portion determination processing is performed. Details of this processing will be described with reference to the flowchart of FIG.

In step S308, an OCR correction determination process is performed. Details of this processing will be described with reference to the flowchart of FIG. When the process of step S308 ends, the process of FIG. 17 ends.
<Description of FIG. 16>

  FIG. 16 is an example of a flowchart of a frame contact portion determination process executed by the client terminal 101 according to the second embodiment of the present invention. Each step of the flowchart of FIG. 16 is executed by the CPU 201 of the client terminal 101.

  In step S1601, for the OCR result performed in step S301 in FIG. 3, rectangle information (also referred to as rectangular coordinates) of the image used for recognition for each digit of the entry box is acquired (a rectangle circumscribing the character). Read the coordinates).

  Specifically, in the OCR process in step S301 in FIG. 3, if the entry character is not in contact with the entry frame, the circumscribed rectangle of the entered character can be acquired (FIG. 18b). On the other hand, when the entry character is in contact with the entry frame, the OCR process in step S301 in FIG. 3 recognizes the position of the entry frame, and the contact position with the entry frame is inside the entry frame. The rectangular coordinates of the character are acquired with the rectangular coordinates (FIG. 18a).

  In step S1602, a search process is performed to determine whether a black pixel exists in the surrounding dots of the rectangular coordinates acquired in step S1601 for each digit of the entry frame.

And the result of the process of step S1602 is recorded on the contact information table of FIG.
If the result of step S1602 is that a black image is present, the presence / absence of contact 1004 in FIG. 10 is present (the entry frame and the entry character are in contact) (a rectangle circumscribing the character read by the reading means). If the coordinates are adjacent to the black pixel, it is determined that the character read by the reading means is replaced). In that case, the number of black pixels in which direction is recorded in the contact presence / absence 1004, the contact location 1005, and the contact ratio 1006 in FIG.

  In step S1602, whether or not to replace the character read by the reading unit based on the character information stored in the storage unit and the overlapping state of the character entry box entered by the user in the present invention. It is an example of the determination means which determines whether or not.

  As the contact ratio in this embodiment, the ratio of how many black pixels exist in the pixels adjacent to the frame line in the entry frame information table in FIG. 6 can be used.

  In step S1603, a process is performed to determine whether a digit (frame) that has not yet been searched for remains in the entry box. If there is a digit that has not yet been searched, the process advances to step S1601. If it does not exist, the frame contact portion determination process is terminated.

  As described above, it is also possible to execute the frame contact portion determination process without using a gray image.

  This completes the description of the OCR result correction process for contact characters.

  In this embodiment, the process of step S308 is executed after step S1701, but as another embodiment, after the process of step S1701 is executed, the characters in the rectangular coordinates acquired in step S1601 are again OCRed. It may be processed and replaced with the recognized character.

  In the above-described two embodiments, the CPU 201 of the client terminal 101 executes the processes of FIGS. 3, 5, 11, and 14-17. However, as another embodiment, the CPU 201 of the server 102 executes each process. May be.

  As described above, according to the present invention, it is possible to provide a mechanism that more accurately corrects OCR misrecognition caused by an entry character overlapping an entry frame.

  Further, according to the present invention, it is possible to efficiently separate the entry characters and the entry frame.

  It should be noted that the configuration and contents of the various data described above are not limited to this, and it goes without saying that the various data and configurations are configured according to the application and purpose.

  The embodiment of the present invention has been described in detail above. However, the present invention can take an embodiment as a system, apparatus, method, program, storage medium, or the like, specifically, a plurality of devices. The present invention may be applied to a system configured from the above, or may be applied to an apparatus including a single device.

  Another object of the present invention is to supply a storage medium storing software program codes for realizing the functions of the above-described embodiments to a system or apparatus, and the computer (or CPU or MPU) of the system or apparatus stores the storage medium. Needless to say, this can also be achieved by reading and executing the program code stored in.

  In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code itself and the storage medium storing the program code constitute the present invention.

  As a storage medium for supplying the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

  Further, by executing the program code read by the computer, not only the functions of the above-described embodiments are realized, but also an OS (basic system or operating system) running on the computer based on the instruction of the program code. Needless to say, a case where the function of the above-described embodiment is realized by performing part or all of the actual processing and the processing is included.

  Further, after the program code read from the storage medium is written in a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function is determined based on the instruction of the program code. It goes without saying that the CPU or the like provided in the expansion board or function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.

  In addition, all the structures which combined each embodiment mentioned above and its modification are also included in this invention.

Network 1
Client terminal 101
Server 102

Claims (9)

Management means for managing predetermined characters in association with other characters;
The document medium with mark frames to fill the character from the data of the image data, reading means for reading the characters entered by the user in the entry boxes,
A character is in a predetermined read in the previous SL reading means, and determining means for determining the said predetermined character read are overlapped with the input frames and a predetermined position,
The determination means determines that the character read by the reading means is the predetermined character, and that the read predetermined character overlaps the entry frame at a predetermined position. And a determining unit that determines to replace the predetermined character read by the reading unit with the other character managed by the managing unit in association with the predetermined character. Information processing device. Output means for outputting the character read by the reading means,
The information processing apparatus according to claim 1, wherein the output unit performs output by replacing the other character determined by the determination unit. A removal means for removing the entry frame included in the image data;
The reading means reads again the characters entered by the user in the entry frame from the image data from which the entry frame has been removed by the removing means,
The determining means determines by the determining means that the character read by the reading means is the predetermined character, and the read predetermined character overlaps the entry frame at a predetermined position. After that, the character obtained by re-reading the predetermined character by the reading unit from the image data from which the entry frame is removed by the removing unit, and managed by the management unit in association with the predetermined character Deciding to replace with the other characters that are
The information processing apparatus according to claim 1, wherein: The predetermined characters are 1, 5, and 7. When the predetermined character is 1, the other character is 7, and when the predetermined character is 5, the other character is 6. 4. The information processing apparatus according to claim 1, wherein when the predetermined character is 7, the other character is 0 or 2. 5. When the predetermined character is 1, the predetermined position is an upper portion of the entry frame, and when the predetermined character is 5, the predetermined position is a left side portion of the entry frame, and the predetermined character is The information processing apparatus according to claim 4, wherein the predetermined position is a left side of the entry frame or a lower part of the entry frame. If the predetermined character is 7 and the predetermined position is the left side of the entry frame, the other character is 0, and the predetermined character is 7, and the predetermined character is The information processing apparatus according to claim 5, wherein the other character is 2 when the position of the character is at a lower part of the entry frame. From the image data, and the density of characters entered by the user, an acquisition unit before acquire the concentration of Kiki Nyuwaku,
And density of characters entered by the user acquired by the acquisition unit, based on the density difference between the density of the input frame, the image data is binarized, a threshold value for only the character to a black pixel Threshold determining means for determining;
With
The removing means uses a threshold determined by the threshold determination means, wherein the image data by the black pixels in only the character by binarizing, wherein characterized that you remove the entry frame Item 7. The information processing device according to any one of Items 3 to 6 . And Jo Tokoro character, a method of controlling an information processing apparatus including a management unit to manage in association with other characters,
Reading means of the information processing apparatus, a document medium example Bei mark frames to fill the character from the data of the image data, the reading step reads a character entered by the user in the entry boxes,
Judging means of the information processing apparatus, wherein a character is a predetermined character read by the reading step, and determining the said predetermined character read Luca not overlap in the entry box and the predetermined position A determination step to
The determining means of the information processing apparatus, in the determination step, the character read in the reading step is the predetermined character, and the read predetermined character overlaps the entry frame at a predetermined position. in response to which has to have been determined, and replacing the predetermined character read by said reading step, and more to the other characters are managed before Symbol management means in association with the predetermined character decision A control method for an information processing apparatus, comprising: a determining step. A predetermined character, a read program executable by an information processing apparatus including a management unit to manage in association with other characters,
The information processing apparatus;
Reading means for reading characters entered by the user in the entry frame from image data obtained by converting a document medium having an entry frame for entering characters into data;
Wherein a character is in a predetermined read by the reading means, and determination means for determining the predetermined character to which the read is Luke not overlap in the entry box and a predetermined position,
Wherein the determining means, wherein a character which is read by the reading means said predetermined character, and, depending on said predetermined character to which the read is determined to be overlapped with the input frames and place Te, the predetermined character read by said reading means, be made to function as a determining means for replacing a decision to the other characters that are more manageable before Symbol management means in association with the predetermined character A program characterized by

Images