JPH0464188A - Method and device for generating document format - Google Patents

Abstract

PURPOSE:To reduce a workload and to prevent the mis-setting of position information performed by setting a read mode in each field in unit of character frame, and converting setting information and each detection information to a document format. CONSTITUTION:A document position is detected by reading out the image data of an inputted document by a document position detecting part 2 after storing in image memory 1 once, and the frame line of the character frame is extracted by a field position detecting part 3, and the position of a field is detected in unit of character frame. Thence, the character in each character frame is recognized at a character recognition part 4, and a read mode setting part 5 sets the read mode based on a recognition result. Finally, the detection information in the document position detecting part 2, the detection information in the field position detecting part 3, and the setting information in the read mode setting part 5 are inputted to an output information conversion part 6, and they are converted to the document formats. Thereby, the workload can be reduced, and also, the mis-setting of the position information can be prevented.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial field of application Conventional technology Problems to be solved by the invention Means for solving the problems (Fig. 1) Working examples (Figs. 2 to 5) Effects of the Invention [Summary] With regard to a form format creation method and a form format creation device, the present invention aims to reduce the amount of work required to create a form format and also prevent mistakes in setting location information. Store the image data in memory, detect the position of the form from this stored image data,
By extracting the character frame line in each field at the detection position, the position of each field is detected in character frame units, and the characters in each detected field are recognized in character frame units. Based on the recognition results, the read mode in each field is set for each character frame, and the corresponding setting information,
The system is configured to create a form format by converting each of the above detected information into a form format.

[Industrial application field]

The present invention relates to a form format creation method and a form format creation device, and more specifically, the present invention is used when recognizing characters on a form using a character recognition device. The present invention relates to a form format creation method and a form format creation device.

[Conventional technology]

Conventionally, forms in various formats have been used for automatic reading by character reading devices.

When reading such a form, it is necessary to create the format of the form in advance. Conventionally, the following method has been used to create a form format.

(1) Fill in the reading information such as the number of fields, field position, number of characters read in the field, reading font (handwritten, printed), field treat mode, etc. of the form to be created on the fixed form for format creation. , created by reading it with a character reading device.

(2) MM I (Man Machine I)
It is created by directly inputting the reading information from (interface).

[Problem to be solved by the invention]

The above-mentioned conventional devices had the following drawbacks.

(1) The reading information to be input is usually written in the manual. Therefore, it is necessary to perform complicated tasks while referring to the manual.

Therefore, it takes a lot of time to input the read information, and the work efficiency is also poor.

(2) Mistakes in measuring and converting positional information such as field positions are likely to occur.

SUMMARY OF THE INVENTION It is an object of the present invention to eliminate such conventional drawbacks, reduce the amount of work required to create a form format, and furthermore prevent errors in setting position information.

[Means to solve the problem]

FIG. 1 is a diagram showing the principle of the present invention, in which 1 is an image memory, 2 is a form position detection section, 3 is a field positioning section, and 4 is a diagram showing the principle of the present invention.
5 indicates a character recognition section, and 5 indicates a read mode setting section.

In order to achieve the above object, the present invention is configured as follows.

(1) Store the image data of the input form in memory, detect the position of the form from this stored image data, and extract the character frame line in each field at the detected position, so that each field The position of the field is detected in each character frame, and the characters in each detected field are recognized in each character frame, and based on this recognition result, the read mode in each field is set in each character frame. , the setting information, and each of the above-mentioned detection information to create a form format by converting the information into a form format.

(2) An image memory 1 that stores the image data of the input form; a form position detection unit 2 that detects the position of the form from the image data in the image memory 1; A field position detection unit 3 detects the position of the field in each character frame by extracting the frame line of the character frame, and a character recognition unit 4 recognizes the characters in each detected field in each character frame. , a read mode in each field is set for each character frame based on the recognition result of the character recognition unit 4, a read mode setting unit 5, setting information of the read mode setting unit 5, and each of the above-mentioned detection information. 1. A form format creation device, comprising: an output information conversion section 6 for converting a document into a form format, and a form format output can be obtained from the output of the output information conversion section 6.

[Effect]

Since the present invention is configured as described above, it has the following effects.

Image data of a form input using an image scanner or the like is stored in the image memory 1. Thereafter, the form position detection section 12 reads out the image data in the image memory 1 and detects the position of the form.

By detecting the position of the form, the size of the form can be determined.

Thereafter, the field position detection unit 3 extracts the frame lines of the character frames using ruled line extraction processing, and detects the position of the field for each character frame.

Next, the character recognition unit 4 recognizes the characters within each detected character frame, and the read mode setting unit 5 sets the read mode based on the recognition result.

Finally, in the output information converter 6, the form position detector 2
, the detection information of the field position detection section 3, and the setting information of the read mode setting section 5 are inputted and converted into a form format.

In this way, a form format can be automatically created simply by writing the font to be recognized within the character frame of the form on which the format is to be created and by inputting the image data of this form.

〔Example〕

Embodiments of the present invention will be described below based on the drawings.

2 to 5 are diagrams showing one embodiment of the present invention, in which FIG. 2 is an example of a form, FIG. 3 is an example of a character field,
Figure 4 is a block diagram of the form format creation device, Figure 5
The figure is a flowchart of the form format creation process.

In the figure, the same reference numerals as in FIG. 1 indicate the same parts. Further, 7 is an image scanner, 8 is a CPU, and 9 is a ruled line extractor.

The form shown in FIG. 2 is used in this embodiment. The number of lines to be processed is 5 lines with line numbers (]) to (5),
Regarding the number of fields in a line, only the line number (1) has two fields, field ■ and field ■, and the other lines have one field.

The number of characters in a line is (1) 9 characters per line, (2) 3 characters per line,
(3) Lines are 4 characters; (4) Lines are 4 characters; (5) Lines are 4 characters.

The number of characters read in the field is 4 for field ■ in line (1)
Characters, (1) line field ■ is 5 characters, (2) line field ■ is 3 characters, (3) line field ■ is 4 characters, (4) line field ■ is 4 characters, (5) line The field ■ is 4 characters long.

Further, as an example of a character field, there is a field as shown in FIG. 3, for example.

In Figure 3 (A), each character frame in the character field has "
This is an example in which all the kana characters for ``Aiueo'' are filled in, and if this is character-recognized, it can be determined that it is a kana field.

In Figure 3 (B), the first two characters are blank spaces and the rest is the number r.
900". In this way, when there is a margin, if the read mode within the field is unified, the read mode of the margin is also made the same. Note that the margin can be specified arbitrarily.

In FIG. 3(C), the first two characters are the kana "Ai" and the remaining three characters are the number "123". In this way, by mixing characters within a field, it is also possible to specify subfield division.

In FIG. 3(D), "A" is a single alphabetic character, "rN" is a single numeric character, "K" is a kana character alone, and the remainder is a blank space (the blank space is aligned with the left character).

In such a case, by specifying read mode in advance for characters that are less likely to be misread, such as alphabets and numbers, it is possible to specify mixed characters.

As the document format creation device, for example, the device shown in FIG. 4 is used.

The image skill 7 is, for example, as shown in Figure 2.
The image scanner is used to read a form to be read, and the image data read from this image scanner is stored in the image memory 1 by the CPU 8. Thereafter, basic information necessary for creating a form format is obtained from the image data stored in the image memory l.

The form position detection unit 2 detects the paper size (form height, form width) by detecting the top, bottom, left and right edges of the image data of the form taken into the image memory 1.

The field position detection unit 3 detects the position of the field by extracting the frame line of the character frame in each field. At this time, regarding the frame line extraction, the ruled line extraction unit 9 performs ruled line extraction processing. .

To extract a character frame, an image scanner capable of detecting a dropout color is used as the image scanner 7 for a form in which a character frame is written in a dropout color.

Through the ruled line extraction processing described above, the coordinates of the end points of F&, ml (left and right frames for character frames) and horizontal lines (upper and lower frames for character frames) on the form are detected.

Furthermore, when determining the field position, the positioning is performed in consideration of the character frames within the field and the spacing between the character frames.

If you set the minimum spacing between fields to be equal to or greater than the maximum spacing between character frames in a field, you can determine the field position (line center position) by knowing the above minimum spacing in advance. , left and right coordinates of each field)
It is possible to detect.

The following information is obtained by detecting the field position by the field position detection section 3.

a. Number of rows to be processed - Know the number of rows from the row center position of each field.

50 Line number - Specify in order from the top of the line.

C8 Number of fields in a row - The coordinates of the row center line are the same,
Specify in order from the left.

d0 Number of characters in line - Detected from the left and right frames of the character frame.

This is either a large frame or a cloth frame. By counting the number of It is possible to

e, Field number - Specify the fields in the line in order from the left.

f, Number of characters read in field - Detected in the same way as Number of characters in line.

g9 Character number - Specify in order from the left side of the character frame in the field.

h. Reading font (handwritten, printed) - differentiated by the size of the character frame, etc.

Next, in the character recognition unit 4, by writing characters in advance in each character frame in the field (see Fig. 3),
Recognize the character.

The read mode setting unit 5 uses the result of the character recognition to set the read mode for each character frame.

The output information conversion unit 6 converts the above-mentioned form position detection information (@sheet size detection information), field position detection information, and read mode setting information into a form format, thereby converting the basic form format. Create parts.

The form format creation process performed by the above device will be explained based on the flowchart shown in FIG. Note that the number of each process is shown in parentheses.

Set the initial information (101) and set the image scanner 7.
Read the form from. The image data of the form read by the image scanner 7 is stored in the image memory 1 (102).

Thereafter, the form position detection unit 2 performs a form position detection process (103), and the position extraction unit 9 extracts the frame line of the character frame (104), and the field position detection unit 3
The field position is detected (105).

Next, each character in the field is extracted in character frame units and character recognition is performed (106).

Using the information of this character recognition result, the read mode setting unit 5 sets the read mode for each character frame (107
). Execute this for all fields (10B)
, the output information of each of the above units is converted into a form format by the output information conversion unit 6 and outputted (109).

Although the embodiments have been described above, the present invention can also be implemented as follows.

(1) If you set the following information before reading the #& vote, you can specify a complex format.

a. If the first field of the form is designated as the D field, the number written in that field is defined as the form format number.

If several types of read modes are set as shown in FIG. 3 before the b0 character is recognized, special specifications can be made.

When creating a form format in which C0 image fields are mixed, by specifying the maximum size of a character frame, character frames larger than that are determined to be image fields.

(2) In the case of a simple form that does not require control information such as data processing, data check, and editing designation terminal, the form format can be created by the processing of the above embodiment.

However, in the case of documents that require control information or small changes (
If it is necessary to change the read mode (such as changing the read mode), or if the read characters are rejected, settings can be made in a conversational manner.

〔Effect of the invention〕

As explained above, the present invention has the following effects.

(1) Write the font you want to recognize within the character frame on the form you want to format, and read the form with an image reader or the like to automatically output the form format.

(2) Therefore, the amount of work is reduced compared to the conventional example, and mistakes in setting position information can be prevented.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the principle of the present invention. FIGS. 2 to 5 are diagrams showing one embodiment of the present invention. FIG. 2 is an example of a form, FIG. 3 is an example of a character field, and FIG. The figure is a block diagram of the form format creation device, and FIG. 5 is a flowchart of the form format creation process. 1-Image memory 2-Form position detection section 3--Field position detection section 4-Character recognition section 5--Read mode setting section 6--Output information conversion section

Claims (2)

[Claims] (1) Store the image data of the input form in memory, detect the position of the form from this stored image data, and extract the character frame line in each field at the detected position. The position of the field is detected in each character frame, and the characters in each detected field are recognized in each character frame.Based on this recognition result, the read mode in each field is set in each character frame. A method for creating a form format, characterized in that a form format is created by converting the setting information and each of the detected information into a form format. (2) An image memory (1) that stores image data of an input form; a form position detection unit (2) that detects the position of the form from the image data in the image memory (2); and the detected form position. The field position detection unit (3) detects the position of the field in each character frame by extracting the frame line of the character frame in each field, and a character recognition unit (4) that recognizes the character; a read mode setting unit (5) that sets the read mode in each field for each character frame based on the recognition result of the character recognition unit (4); An output information converting section (6) is provided to convert the setting information of the mode setting section (5) and each of the above detection information into a form format, and the form format output is determined from the output of the output information converting section (6). A document format creation device characterized in that it can obtain the following information.