JPH0765111A - Optical character recognition device - Google Patents

Abstract

(57) [Abstract] [Purpose] To provide an optical character recognizing device which can easily perform a designation operation for reading a character and has a small load associated with data processing after character recognition. [Arrangement] This apparatus has a layout information creation unit 70, an output order information creation unit 71, a storage unit 6, and a character recognition unit 3. The layout information and the output sequence information can be created by the designation operation, and when the form identification information is designated, the character recognition unit 3
Outputs recognition result data corresponding to each item in the designated item order. As a result, it is possible to obtain the recognition result data in which the output order is unified regardless of the actual item order of the form to be processed, though the specifying operation is easy.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION The present invention relates to an optical character recognition device (OCR).

[0002]

2. Description of the Related Art The methods of recognizing characters written on a form by OCR are roughly classified into the following three methods.

The first method is a table (F) in which the coordinates of the reading area (field), the number of digits to be read, the type of character, etc. are designated in advance.
(C information) is created in advance, and OC is performed based on the FC information.
R is a method for recognizing characters. In the second method, the layout information (LC information) of the reading area is created in advance,
The OCR determines the reading area based on the LC information,
This is a method of recognizing characters. The third method is a method in which the OCR automatically recognizes a reading area from a form image obtained by optical scanning and recognizes a character. The OCR according to the first method is called a form OCR, and the OCRs according to the second and third methods are called a text reader.

In the form OCR, since the recognition result is obtained as the coordinates of the reading area are specified, a variety of functions such as checking, editing, and calculation can be realized, and it is easy to know what item the recognition result relates to. However, it has the advantage that the recognition result data can be handled easily.
Since it is necessary to strictly specify each kind of specification, there is a drawback that it takes time and effort to specify.

On the other hand, the text reader has an advantage that the layout can be easily specified even in the case of the second method, but it is difficult to judge which item the recognition result data corresponds to. Have. For this reason, the load of data processing after character recognition increases.

[0006]

Therefore, for example, when a plurality of forms having different formats such as a transfer request form are totaled, the arrangement order of each item such as a bank name, a branch name, an account number and a payee name is arranged. However, the form OCR has a problem that it takes a lot of time to specify, and the text reader has a problem that it is not easy to process the recognition result data.

Therefore, the present invention has been made in view of the above circumstances, and provides an optical character recognition device which is easy to perform a designation operation for reading a character but has a small load associated with data processing after character recognition. The purpose is to provide.

[0008]

In order to achieve the above object, an optical character recognition device according to a first aspect of the present invention is a form in which reading areas are provided corresponding to a plurality of items arranged in an arbitrary order. In the optical character recognition device that reads the characters entered in each reading area of the form and outputs the recognition result data, select some or all of the plurality of items, and The present invention is characterized by having a specifying means for specifying an item order and an output control means for outputting recognition result data corresponding to each item in accordance with the item order specified by the specifying means.

The optical character recognition device according to a second aspect of the present invention processes a form in which reading areas are provided corresponding to a plurality of items arranged in an arbitrary order, and the reading areas of the form are processed. In the optical character recognition device for reading the character entered in and outputting the recognition result data, a layout information creation unit for creating layout information of the reading area based on a designation operation, and the plurality of layout information based on the designation operation. An output order information creation unit that creates output order information of each recognition result data corresponding to some or all of the selected items, and a storage that stores the layout information and the output order information together with the corresponding form identification information. Section, layout information and output order information corresponding to the specified form identification information are read from the storage section, and the read layout information and And a character recognition unit for outputting recognition result data corresponding to each item according to a specified item order after performing a recognition process on the characters written in each reading area based on the force order information. It is a thing.

An optical character recognition apparatus according to a third aspect of the present invention is characterized in that it has a list creation unit for creating a list of the form identification information for which the layout information and output order information have been created. is there.

[0011]

According to the optical character recognition device of claim 1,
When an item is selected by the designating means and the item order is designated, the output control means outputs the recognition result data corresponding to each item in accordance with the designated item order. As a result, it is possible to obtain the recognition result data in which the output order is unified regardless of the actual item order of the form to be processed, though the specifying operation is easy.

According to the optical character recognition device of the second aspect, the layout information and the output sequence information can be created by the designation operation, and when the form identification information is designated, the character recognition unit
The recognition result data corresponding to each item is output according to the specified item order. As a result, it is possible to obtain the recognition result data in which the output order is unified regardless of the actual item order of the form to be processed, though the specifying operation is easy.

According to the optical character recognition device of the third aspect, it becomes easy to judge whether or not the layout information and the output sequence information have been created based on the list, and the burden on the operator can be reduced.

[0014]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 1 is a schematic block diagram showing an embodiment of the optical character recognition device of the present invention.

The apparatus of this embodiment shown in FIG. 1 includes a scanner unit 1 for detecting an image 11 of a form 10, an image buffer 2 for storing the form image 11, and character recognition for recognizing characters written on the form 10. A unit 3, a display unit 4 for displaying the detected form image 11 and the like, an input unit 5 for performing various designation operations and input operations, and a storage unit 6 for storing layout information and output order information described below. The control unit 7 controls each unit of this apparatus.

Next, the details of each of the above parts will be described.

The scanner unit 1 includes a photoelectric conversion element, and optically scans the form 10 to detect the form image 11.

The character recognizing unit 3 includes an image buffer 2
A character image of each reading area (see FIG. 2) 13 and the item 12 is cut out from the form image 11 stored on the basis of layout information described later, the similarity is obtained by collation with a character recognition dictionary, and the similarity is calculated. The higher one is output as the recognition result data. The recognition result data is output according to the item order information described later.

The display unit 4 displays the form image 11 stored in the image buffer 2 and guidance information for operation for the operator in an interactive form.

The input section 5 is provided with a mouse, a keyboard and the like, and a designated operation necessary for recognizing characters on the form 10, for example, a layout of the reading area 13 provided corresponding to each item 12 on the form 10. The designation operation necessary for data processing after the recognition result, for example, the output order of each recognition result data is designated. In addition, the input unit 5 can select various modes such as a definition mode and a reading mode with a keyboard, and when the definition mode or the reading mode is selected, identification information (client code) of the form 10 is selected. You can enter.

The storage section 6 comprises a layout information storage section 60 for storing layout information and an output order information storage section 61 for storing output order information.

The control section 7 includes a layout information creating section 7
0 and an output order information creation unit 71 are provided. The layout information creation unit 70 creates layout information based on a designation operation on the input unit 5. The output sequence information creation unit 71 uses the input operation of the input unit 5 to specify the item 1
The second output order information is created. The created layout information and output order information are stored in the layout information storage unit 60 and the output order information storage unit 61 of the storage unit 6 together with the identification information (client code) for specifying the format of the form 10. There is. In this embodiment, the transfer request form is to be read, and the format of this transfer request form has been created on the client side.
Since the format of the form 10 can be specified by specifying the client code, the client code is used as the identification information. Note that this identification information may be other information as long as the information can identify the format of the form 10.

Next, the operation of this embodiment will be described with reference to the flow charts shown in FIGS. The case where the form 10 shown in FIG. 2 is the reading target will be described. On the form 10 to be read, as shown in the figure, a plurality of bank names (upper row) 12a, branch names (lower row) 12b, deposit items 12c, account numbers 12d, and recipients 12e arranged in any order. It is assumed that each item 12 such as the amount of money 12f is provided, a reading area 13 is provided corresponding to each item 12, and a character is written in each reading area 13.

First, it is assumed that the format of the form 10 shown in FIG. 2 has not been registered.

First, the operator operates the keyboard of the input section 5 to select the definition mode and inputs the client code which is the identification information of the form 10 (S1).

Under the control of the control unit 7, the scanner unit 1
The image 11 of the new format form 10 is detected (S
2) The form image 11 is stored in the image buffer 2.

The control unit 7 displays the form image 11 stored in the image buffer 2 on the display unit 4 (S3).
An example thereof is shown in FIG. As shown in the figure, the form image 11 detected by the scanner unit 1 is displayed on the left side of the display screen as black and white binary data. By operating the input unit 5, the form image 11 on the left side of the screen can be partially enlarged and displayed on the right side of the screen as shown in FIG.

Next, as shown in FIG. 4, the form image 1 displayed on the left side of the screen is operated by operating the mouse of the input unit 5.
In order to select the item 12 that requires the recognition result data among the plurality of items 12a to 12f on the first item, a rectangular designated region 14 is set in the reading region 13 corresponding to each required item 12 Then, the control unit 7 automatically sets the field name 15 (S4). In the figure, the designated area 1 is included in the reading area 13 for the bank name 12a and the branch name 12b.
4 is set and "AA-2" is automatically set as the field name 15. When selecting other items 12 other than the bank name 12a and the branch name 12b, for example, as shown in FIG.
The item 12 of e is selected and the same is done for these.
In this way, the layout of the reading area 13 provided corresponding to the plurality of items 12 is designated.

Subsequently, the layout information creation section 7 of the control section 7
0 creates layout information based on the layout of the reading area 13 designated by the input unit 5 (S5), and the form 1
It is stored in the layout information storage unit 60 together with the identification information of 0 (client code).

Next, as shown in FIG. 6, the operator operates the mouse of the input unit 5 to move the mouse into the reading area 13 to be first output on the form image 11 displayed on the left side of the screen. After moving the pointer 16 and clicking,
By moving and clicking the mouse pointer 16 in the reading area 13 to be output next, and sequentially operating in this manner, the output order of the recognition result data of the reading area 13 provided corresponding to each item 12 is changed. Designate (S6). On the right side of the screen, as shown in the figure, the field names (AA-1, AA-2, ...) 15 already named are displayed in the output order.

Subsequently, the output sequence information creation unit 71 of the control unit 7
Creates output order information based on the output order specified by the input unit 5 (S7) and stores it in the output order storage unit 61 together with the identification information (client code).

In this way, the reading area 13 of the form 10 is read.
And the output order of the item 12 are defined.

Next, the case of reading the defined form 10 will be described.

First, the operator operates the keyboard of the input unit 5 to select the reading mode, inputs the registered client code as the identification information of the form 10, and ENDs the input unit 5.
Press the key. (S10).

Based on the depression of the END key, the control section 7
The reading of the form 10 is started. When the operator presses the ESC key of the input unit 5, the control unit 7 causes the form definition list (requester information) that is a list of form identification information for which layout information and output order information have been created. ) Is displayed on the display unit 4.

Under the control of the control unit 7, the scanner unit 1
The image 11 of the registered form 10 is detected (S11),
The form image 11 is stored in the image buffer 2.

The control unit 7 reads layout information corresponding to the identification information (client code) input by the input unit 5 from the layout information storage unit 60, and based on the layout information, each reading area from the form image 11. The layout of 13 is analyzed.

The character recognition unit 3 recognizes the characters written in the reading area 13 and the characters of the item 12 using the layout analysis result by the control unit 7.

The control unit 7 outputs the output sequence information corresponding to the client code input by the input unit 5 to the output sequence storage unit 61.
7, and outputs the recognition result data A1, B1, C1, A2, ... By the character recognition unit 3 to the control unit 7 in accordance with the read output sequence information. That is, the recognition result data A1, B1, C1 for the first row is output according to the order of the bank name 12a, the branch name 12b, the account number 12d, and the recipient 12e of the designated items, and then the recognition result data A2 for the next row. , B2, C2 are output, and the recognition result data is cyclically output to the subsequent lines in the same order as the specified item order.

Here, when the operator wants to display the output result to the control unit 7 for confirmation or correction of the recognition result, he / she operates the input unit 5 to display it as shown in FIG. A correction screen is displayed on the display screen of the section 4. The figure shows a state in which a part of the recognition result data is displayed according to the designated item order such as the bank name 12a, the branch name 12b, the account number 12d, and the recipient 12e. The hatched area in the figure shows character images, and the recognition result data is displayed below each character image. Further, "?" In the figure means a symbol indicating that it was rejected because it could not be recognized, or even if it could be recognized, it could be misread. When "?" Is displayed, the mouse pointer 16 is moved to that position and a correct character is input from the keyboard of the input unit 5, whereby the character is converted into correct character data.

According to the above-described embodiment, the layout of the reading area 13 can be designated only by setting the designated area 14 in each reading area 13 on the display screen of the form image 12, like the form OCR. Since it is not necessary to strictly specify the coordinates and the like, the specification operation for reading the character can be easily performed. In addition, after recognition, the recognition result data corresponding to each item is output according to the specified output order, so even when tabulating different formats such as a transfer request form, the output result with a uniform output order Can be obtained, and the burden on the operator can be reduced in the subsequent data processing such as tabulation.

The present invention is not limited to the above embodiment,
Various modifications can be made without changing the gist of the invention.
For example, although the item 12 that requires the recognition result data is specified when the layout of the reading area 13 is specified, the required item 12 may be specified when the item order is specified.

[0044]

According to the present invention described in detail above, the following effects can be obtained.

According to the first aspect of the present invention, since the recognition result data corresponding to each item is output in the designated item order, the designation operation becomes easy. Further, regardless of the actual item order of the form to be processed, it is possible to obtain the recognition result data in which the output order is unified, so that the burden of data processing after character recognition can be reduced.

According to the second aspect of the invention, the layout information and the output order information can be created by the designation operation, and the recognition result data in which the output order is unified can be obtained even if the forms have different item orders. Since it is possible to specify the character, it is possible to reduce the burden of data processing after character recognition, while making it easy to specify the character.

According to the third aspect of the present invention, it becomes easy to determine whether the layout information and the output order information have been created from the list, and the burden on the operator can be reduced.

[Brief description of drawings]

FIG. 1 is a schematic block diagram showing an embodiment of an optical character recognition device of the present invention.

FIG. 2 is a diagram showing an example of a form.

FIG. 3 is a diagram showing an example of a display screen of recognition result data.

FIG. 4 is a diagram showing an example of a display screen for explaining a method of designating a layout of a reading area.

FIG. 5 is a diagram showing an example of a display screen for explaining a method of designating a layout of a reading area.

FIG. 6 is a diagram showing an example of a display screen for explaining a method of designating an output order of recognition result data of a reading area provided corresponding to each item.

FIG. 7 is a diagram showing an example of output of recognition result data.

FIG. 8 is a diagram showing an example of a correction screen.

FIG. 9 is a flowchart showing the operation of this embodiment.

FIG. 10 is a flowchart showing the operation of this embodiment.

[Explanation of symbols]

 3 character recognition unit 5 input unit 6 storage unit 60 layout information storage unit 61 output order information storage unit 7 control unit 70 layout information creation unit 71 output order information creation unit 10 forms 12, 12a to 12f items 13 reading area

Claims (3)

[Claims] 1. A processing target is a form in which a reading area is provided corresponding to a plurality of items arranged in an arbitrary order, characters read in each reading area of the form are read, and recognition result data thereof is read. In the optical character recognition device for outputting, a part of or all of the plurality of items are selected, and a specifying means for specifying the item order, and each item according to the item order specified by the specifying means. An optical character recognition device, comprising: output control means for outputting corresponding recognition result data. 2. A form in which a reading area is provided corresponding to a plurality of items arranged in an arbitrary order is processed, a character written in each reading area of the form is read, and recognition result data thereof is read. In the optical character recognition device for outputting, a layout information creation unit that creates layout information of the reading area based on a designated operation, and some or all of the selected items of the plurality of items based on the designated operation. Output order information creation unit that creates output order information of each recognition result data corresponding to, a storage unit that stores the layout information and output order information together with corresponding form identification information, and corresponds to the specified form identification information. The layout information and the output order information are read from the storage unit, and are written in each reading area based on the read layout information and the output order information. After the recognition processing for the characters, the optical character recognition apparatus characterized by comprising: a character recognition unit which outputs the recognition result data corresponding to each item according to the specified item order, the. 3. The optical character recognition device according to claim 1, further comprising a list creation unit that creates a list of the form identification information for which the layout information and the output order information have been created.