JPH0272497A - Optical character reader - Google Patents

Abstract

PURPOSE:To surely read a character string even when dirt, etc., are exist by setting a character string extraction designating code corresponding to a read field as FC (format control) data for the character string data obtained as the result of a character recognizing processing. CONSTITUTION:A character string memory part 16, which stores character string data obtained as the result of the recognition of the character recorded on the read field, and a character string extraction designating information setting means, which sets character string extraction designating information to indicate the constitution of the character string to be outputted as the result of the recognition of the read field in the FC data, are provided. Based on the character string extraction designating information stored into an FC data memory part 15, the character data are extracted from the character string data stored into the character string memory part 16, and the character string data obtained as the result of it are outputted as a final recognition result. Thus, even when the dirt, etc., exist in the read field, the character string recorded in the read field can be surely read.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to an optical character reading device that performs reading processing on characters recorded in a reading field based on format control data.

(Prior Art) Generally, a document that is read by an optical character reading device is provided with a reading field that indicates an area where characters are recorded. The optical character reading device performs character reading processing on characters recorded in a reading field based on format control data (hereinafter abbreviated as FC data). The FC data includes information such as the position of the reading field, the type of characters to be recorded, the character font, and the character pitch.

By the way, around the reading field provided on the form, there is provided an area called a clear area that prohibits the recording of characters and the presence of stains. This is because, if dirt or the like is present in the reading field, the optical character reading device performs recognition processing by regarding the dirt or the like as a character recorded within the field.

For example, if dirt or the like exists in the clear area or reading field, recognition processing for the dirt or the like will be performed based on the FC data. As a result, misreading, refusal to read, etc. occur, making it difficult to output accurate data.

(Problem to be Solved by the Invention) In this way, in the conventional optical character reading device, if dirt or the like is present in the reading field, recognition processing within the reading field indicated by the FC data is performed even for the dirt. It ends up. As a result, misreading, refusal to read, etc. may occur, and accurate data may not be output.

The present invention has been made in view of the above points, and is an optical system that can reliably read the character string recorded in the reading field even if there is dirt or the like in the reading field. The purpose is to provide a character reading device.

[Structure of the Invention] (Means for Solving the Problems) The present invention provides an optical character reading system that performs reading processing based on FC data for characters recorded in a reading field provided in a form to be read. In the device, a character string storage means for storing character string data obtained as a result of recognizing the characters recorded in the reading field, and characters indicating the structure of the character string to be output as a result of recognizing the reading field in the FC data. and a string extraction specification information setting means for setting column extraction specification information, from the string data stored in the string storage means, based on the string extraction specification information set by the string extraction specification information setting means. The system is configured to extract character data and output the resulting character string data as the final recognition result.

(Function) In the optical character reading device constructed in this way, when the contents of the character string recorded in the reading field are known to some extent in advance, a character string extraction method for specifying the character string can be used. By setting the fixed code as FC data, only the desired character string can be extracted and output.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an optical character reading device according to an embodiment of the present invention.

The optical character reading device shown in FIG.
Under the control of the computer, the characters recorded on the form are read. The external control device IO is a device that controls the entire system and gives instructions to the control unit 11. Control unit lO
In accordance with instructions given from the external control device 10, the reading process of the characters recorded on the form is performed by controlling each part. The form transport mechanism section 12 passes a form to be read under the control of the control section 11 over a reading line where optical scanning by the photoelectric conversion section 13 is performed. The photoelectric conversion unit 13 irradiates light on the reading line to the form conveyed by the form conveyance mechanism unit 12 . The photoelectric conversion unit 13 performs photoelectric conversion on the light reflected by the form. The electrical signal obtained by this photoelectric conversion is quantized based on a preset reference value. The character recognition unit 14 stores the quantized form image data in, for example, a sheet buffer (not shown). The character recognition unit 14 detects and extracts character patterns from the form image data stored in the sheet buffer based on the FC data held in the FC data storage unit 15, and performs character recognition processing for each character. The FC data stored in the FCC data storage unit 15 includes the position of a reading field set in advance on a form and the character font of the characters recorded in this reading field.

Information indicating the character type, character pitch, etc. is included.

Furthermore, a character string extraction designation code for further extracting character data from the character string data recognized and obtained by the character recognition unit 14 is stored.

The recognition result recognized by the character recognition unit 14 is temporarily stored in the character string storage unit ta. The control unit 11 performs character string extraction processing and character string editing on the character string data stored in the character string storage unit IG based on the character string extraction designation code stored in the FC data storage unit 15. Do this. The character string data obtained through this character string extraction process is output to the external control device 10 as the final recognition result of the characters recorded in the reading field.

Next, the operation of this embodiment will be explained.

Here, a document 20 having a format as shown in FIG. 2 is to be read. In addition, the character string r01234J is recorded in the reading field 21 provided in the form 20, and 1
It is assumed that 9 exists.

First, the form 20 is conveyed by the form conveyance mechanism section 12, and is conveyed on the reading line of the photoelectric conversion section 13 from the leading end to the trailing end of the form 20. As a result, a form image for one form is stored in the sheet buffer of the first literary magazine 14. The control unit 11 detects and extracts character pattern data for each character based on the FC data corresponding to the read field 21 stored in the FC data storage unit 15. Here, the reading field 21 indicating the position to be subjected to character recognition processing is specified by the FC data.

That is, a line is indicated by a distance ML from the leading edge of the form 20, and the starting and ending ends of the reading field are indicated by distances MS and ME with the left end of the form 20 as a reference. Further, when performing recognition processing for characters in the reading field 21, the recognition processing is performed based on data such as a three-character character type font stored as FC data. The data obtained as a result of this recognition is stored in the character string storage section 16 as character string data recorded in the reading field 21. Character string data obtained as a result of recognition of this reading field 21 is shown in FIG. string r01234
J is accurately recognized as shown in FIG. 3, and the recognition process is performed by treating the dirt on both sides of the character string in the reading field 21 as two characters.

As a result, the dirty portion is refused to be read, and data "?" indicating that the reading is refused is added to the character string data.

Next, the control unit 11 performs search 1 editing on the character string data stored in the character string storage unit based on the character string extraction designation code stored in the FC data storage unit 15.

Here, the character string extraction specification code will be explained. The character string extraction designation code is set as one of the FC data, and is set for each reading field. A specific example of the character string extraction specification code and its extraction operation is shown below.

X'FF' . . . The result recognized by the character recognition unit 14 is output as is.

X'　FO'　・・・Extract only the specified digit.

However, the type of characters to be output is not specified.

0-9, A-Z...Applicable only to the specified digit, the specified character and the reading result are compared, and the reading result is output only when they match.

Here, an example of an output result corresponding to the case where the character string extraction designation code for the reading field 21 is set as shown in (1) to (4) shown in FIG. 4 will be described.

In the case of the extraction character string specification (X'FF' ”) as shown in (1), no extraction operation is performed, so the character string data r?? stored in the character string storage unit 16 is
o1234? ? Output J as is.

In the case of extraction character string specification (X'FO'...X'FO") as shown in (2), the character string data r?
Output.

In the case of extraction character string specification (0, has been done. Therefore, a search is performed to see if data with the number "0" exists from the beginning of the character string data. As a result, the character data in the third digit of the character string data match. Then, since the extraction operation is not performed after the data "0", the character string data r01234? ? Output J.

In the extraction character string specification (0, X'FO' ... Column data output is specified. For this reason,
A search is performed to see if data with the number "0" exists from the beginning of the character string data. As a result, the character data in the third digit of the character string data match. Therefore, the character data indicating the number "0" and the subsequent 4-digit character string data r012
Outputs 34J.

In this way, an output character string can be specified for the character string data recognized by the character recognition unit 14 based on the character string extraction designation code.

Therefore, the character string ro recorded in the reading field 21
For 1234J, if the number of characters (5), the first character rOJ, etc. are known to some extent, accurate recognition results can be obtained by specifying the output character string. Therefore, by specifying the extracted character string as shown in (4) in FIG. It is possible to accurately extract and output only the character string that is specified.

[Effects of the Invention] As described above, according to the present invention, regarding character string data obtained as a result of character recognition processing for a reading field,
By setting the character string extraction designation code corresponding to the reading field as FC data in advance, the desired character string can be extracted. Therefore, even if there is dirt or the like in the reading field, it is possible to reliably read the character string.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the structure of an optical character reading device according to an embodiment of the present invention, FIG. 2 is a block diagram showing the structure of a read form used in the same embodiment, and FIG. FIG. 4 is a diagram showing character strings obtained as a result of recognition processing of the reading field of the shown form, and FIG. 4 is a diagram showing an example of setting a character string extraction designation code and the corresponding output result. lO...External control device, 11...Control unit, 12...
- Form transport mechanism section, 13... Photoelectric conversion section, 14...
Character recognition unit, 15...FC data storage unit, 1G...
String storage. Figure 2 Figure 3 Figure 1 Figure 4

Claims (1)

[Scope of Claims] An optical character reading device that performs reading processing based on format control data on characters recorded in a reading field provided in a form to be read, comprising: a character string storage means for storing character string data obtained as a result of recognizing the reading field; and character string extraction specification information indicating a structure of a character string to be output as a result of recognizing the reading field is set in the format control data. a character string extraction designation information setting means for extracting character data from the character string data stored in the character string storage means based on the character string extraction designation information set by the character string extraction designation information setting means; 1. An optical character reading device comprising: an extraction processing means.