JPH06162255A - Character reading system - Google Patents

Abstract

PURPOSE:To provide a character read technique which can realize a reduction of an error correction work by an operator. CONSTITUTION:Under the control of a main control part 7, a business form 3 is converted to binarized data by a scanner part 4 and fetched, and with regard to each field, character recognition is executed in a recognition control part 6. As for a field in which an error is generated at the time of character recognition, whether the field concerned is a specific read field which dispenses with read accuracy of a high level or not is checked by referring to a recognition accuracy level setting part 5a, and in the case of the specific read field, a result of character recognition is delivered to a conversion control part 5. With regard to the specific read field, the conversion control part 5 executes automatically a correction of an abnormal digit, and forced substitution by a candidate character of an unreadable character and a specific conversion character without allowing an operator to intervene, and thereafter, reports a result of conversion to a host device 2 through a transmission control part 8.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character reading technique, and more particularly to a technique effective when applied to a character recognition process in a form in which a plurality of types of reading fields having different required reading accuracy are mixed.

[0002]

2. Description of the Related Art For example, in information processing systems and the like, as a means of data input, a known optical character recognition (OCR) technique is used to automatically read handwritten characters from a medium such as a form. There is. In such a character reading technique, it is usually necessary to correct unreadable characters.

Conventionally, as such a character reading technique, as disclosed in, for example, Japanese Patent Application Laid-Open No. 61-163472, the operator displays all the character patterns of the field to which the unrecognizable character belongs on the display. In order to enable an operator to accurately correct an unrecognizable character from a keyboard by referring to another readable character in a field, a technique is known.

[0004]

However, in the case of the above-mentioned prior art, no consideration is given to the type of field and, for example, other fields such as a remarks column that does not require highly accurate reading are also considered. In the same manner as the field in which important information such as the amount of money is written, there is a problem that the correction operation of the unreadable character is required uniformly by the intervention of the operator, and the correction operation becomes complicated.

An object of the present invention is to provide a character reading technique capable of reducing the error correction work by the operator.

The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

[0007]

Of the devices disclosed in the present application, representative ones will be briefly described as follows.
It is as follows.

According to the present invention, there is provided a reading accuracy discriminating means for designating a specific reading field which does not require a high reading accuracy in a character reading system in which a desired format pattern is specified by a scanner by reading a desired form pattern and character recognition is performed. , A first operation for forcibly replacing an unread character with a candidate character or a specific replacement character when a character recognition error occurs in the specific reading field, which consists of at least one of an unread character and an abnormal digit exceeding the specified number of characters And an unreadable character conversion means for performing at least one of the second operations for correcting the abnormal digit by truncating the characters exceeding the designated number of characters.

[0009]

According to the character reading technique of the present invention described above,
For a specific reading field that does not require a high degree of reading accuracy, which is specified by the reading accuracy discrimination means, the unread character conversion means forces replacement with unread character candidate characters or specific replacement characters, and abnormal digits. Since error corrections such as truncation of excess characters are automatically performed, operator intervention is not required when modifying specific reading fields that do not require a high degree of reading accuracy, such as in the remarks column. The burden on the operator can be reliably reduced. Further, the time required for the correction work can be shortened.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A character reading method according to an embodiment of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram showing an example of the configuration of an optical character reading device 1 to which the character reading method of this embodiment is applied.

The optical character reader 1 receives a read command from the host device 2 via the transmission controller 8 and, under the control of the main controller 7, scans the image of the form 3 into the scanner unit 4.
At, the form image data is captured after being converted into a binary pattern. The recognition control unit 6 performs character recognition processing on the captured form image data.

In the case of the present embodiment, when a reading error occurs in the character recognition processing in the recognition control section 6, the conversion control section 5 is activated as necessary. The conversion control unit 5 determines whether or not the field is a conversion target field specified in the form format by referring to the information set in the recognition accuracy level setting unit 5a.

That is, in the case of the present embodiment, the recognition accuracy level setting unit 5a determines whether or not a high reading accuracy is required for each field as the form format information, the maximum number of characters allowed in the field, Information such as what character is set as the conversion character 10 used in the conversion processing described later is specified.

Then, the field is a conversion target field that does not require a high degree of reading accuracy and may be automatically subjected to error correction, and an unreadable character is designated in the field as a character recognition result. If there is an abnormal digit exceeding the number of characters, etc., the excess characters are truncated to eliminate the abnormal digit, and then the unread character is converted to a candidate character or a specific replacement character if there is no candidate character. And the conversion result is reported to the main control unit 7. The main control unit 7 sends the conversion result to the higher-level device 2 via the transmission control unit 8.

An example of the operation of the character reading method of this embodiment will be described below.

FIG. 2 is a conceptual diagram showing an example of conversion processing of unread characters in the conversion control unit 5, and FIGS.
3 is a flow chart showing an example of the operation of the character reading method of the present embodiment.

First, the scanner unit 4 takes in the form image data 9 (step 21) and cuts out a field based on a specified form format (step 2).
2) Character recognition is performed from the field (step 2)
3).

Then, it is determined whether or not there is a recognition error (step 24), and if there is an error, it is further determined whether or not high-precision recognition is not required for the field (step 25).

If high-accuracy reading is required, such as in the amount column, that is, if there is no designation that high-accuracy recognition is not required, error occurrence information is set (step 27), and the error occurrence occurs. The character recognition result 11 is directly output to the upper level device 2 together with the information (step 2
8) Check to see if it is the last field (step 2
9) Step 2 if there are unprocessed fields
Repeat from 2 onwards.

On the other hand, in step 25, if the field in which the error has occurred is, for example, in the remarks column and there is a designation that high-precision recognition is not required, the conversion control unit 5 performs a correction process (step 26).

That is, in this step 26, as illustrated in FIG. 4, it is first checked whether or not the number of characters of the character recognition result 11 is within the limited number of characters specified in the field (step 26a), and the number of characters is determined. In the case of an excessive abnormal digit, the characters overflowing from the character recognition result 11 are truncated (step 26b).

Thereafter, an unread character is searched for in the remaining character recognition result 11 (step 26c), and it is checked whether or not there is a candidate character for the found unread character (step 26d). , Replace unread characters with candidate characters (step 26e). If no candidate character exists, the unread character is forcibly replaced by the specific conversion character 10 (in this embodiment, "@" is set as an example) (step 2).
6f).

The above steps 26c to 26f are repeated until there are no unread characters (step 26
g), the conversion result 12 is reported to the main controller 7 (step 26h).

As for the specific conversion character 10, an arbitrary character may be assigned to each field by adding information about the conversion character 10 to a part of the form format information.

Although the invention made by the present inventor has been specifically described based on the embodiments, the invention is not limited to the embodiments and various modifications can be made without departing from the scope of the invention. Needless to say.

For example, in the above description of the embodiments, the case where the character reading system of the present invention is adopted in the optical character reading device has been described as an example, but the present invention is not limited to this.
The present invention can be adopted as long as it is a character reading system that character-recognizes a form specified in the form format and outputs the character recognition result and a candidate character corresponding to an unreadable character and an abnormal digit. For example, if the optical character reading device is a device having the above-mentioned character reading method, an unread character can be automatically converted by a higher-level device, and the object of the present invention can be achieved.

[0028]

The effects obtained by the typical ones of the inventions disclosed in the present application will be briefly described as follows.
It is as follows.

That is, according to the character reading method of the present invention, regarding a specific reading field such as a remarks column that does not require a high reading accuracy, correction of abnormal digits by truncation of excess characters, and further unread character candidates. Characters and specific conversion characters are automatically replaced, reducing error correction work by the operator, for example, in character reading processing of a form that contains multiple types of reading fields with different required character reading accuracy. The effect that can be realized is obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of a configuration of an optical character reading device to which a character reading method according to an embodiment of the present invention is applied.

FIG. 2 is a conceptual diagram showing an example of conversion processing of an unread character in a conversion control unit.

FIG. 3 is a flowchart showing an example of the operation of the character reading method according to the embodiment of the present invention.

FIG. 4 is a flowchart showing an example of the operation of the character reading method according to the embodiment of the present invention.

[Explanation of symbols]

 1 Optical Character Reading Device 2 Upper Device 3 Form 4 Scanner Unit 5 Conversion Control Unit (Unreadable Character Conversion Unit) 5a Recognition Accuracy Level Setting Unit (Reading Accuracy Discriminating Means) 6 Recognition Control Unit 7 Main Control Unit 8 Transmission Control Unit 9 Form Image data 10 Converted character 11 Character recognition result 12 Converted result

Claims (1)

[Claims] 1. A reading precision discriminating means for designating a form format and reading a desired form pattern with a scanner to perform character recognition, and a reading precision discriminating means for designating a specific reading field which does not require a high reading precision. A forcibly replacing the unread character with a candidate character or a specific replacement character when a character recognition error is generated in the specific reading field, the character recognition error including at least one of an unread character and an abnormal digit exceeding a specified number of characters And a non-reading character conversion means for performing at least one of the second operation for correcting the abnormal digit by truncating the characters exceeding the specified number of characters.