JP2639920B2 - Character recognition device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Field of Industrial Use In the present invention, when a predetermined format, for example, a plurality of data described in a slip is treated as a data block, a delimiter marks the delimiters between the data blocks. The present invention also relates to a character recognition device for sequentially recognizing characters described for each data block.

(B) Conventional technology Generally, accounting slips (hereinafter abbreviated as slips) used by various departments such as companies or organizations and public institutions are composed of a single sheet,
Since it is a single slip, in an optical character reading device called OCR, changing the paper was equivalent to changing the slip.

Therefore, instead of treating as a single-sheet form, if a plurality of form contents are entered on a long continuous form and the OCR recognizes the form, even if the number of data (the amount of data) of each form is different, the form It is necessary to determine the breaks where the content, ie the data blocks, are distinct.

A conventional page break as a break is disclosed in Japanese Patent Application No. 60-128789 filed by the present applicant.

(C) Problems to be Solved by the Invention Conventionally, as described above, detection of a page break has been proposed. However, when the recognition processing of the plurality of slips is performed continuously, the contents of the slips are printed on continuous paper. It is not possible to solve the problem of efficient processing after filling in.

The present invention has no problem if, for example, a large number of slips to be processed in one company can be sequentially keyed in from a terminal device by an employee of the same company into a host computer, but an enormous number of slips are involved in this method. Acquisition and costs within one company are enormous.

(D) Means for Solving the Problems The present invention has a paper feed unit for conveying an input sheet, and reads characters written on the input sheet by a character observation unit.
In a character recognition device for recognizing the characters, the output is added to a recognition unit, and a character representing a plurality of data based on a predetermined format is described on a continuous paper used as the input paper, and a delimiter mark is provided between the plurality of data. A specific character or symbol is described in advance, and a recognition output derived through the character observation unit and the recognition unit is added to an output processing unit, and the output processing unit detects a data block unit break,
In this configuration, characters described for each data block are sequentially recognized.

(E) Function In the present invention, predetermined characters or symbols are written between data blocks of continuous paper used as input paper, these are read by a character observation unit, and a recognition output is added to an output processing unit via a recognition unit. The predetermined character or symbol can be detected as a break between data blocks.

(F) Embodiment The present invention will be described with reference to the drawings. FIG. 1 is a system block diagram of a character recognition device of the present invention, and FIG. 2 is an explanatory diagram showing input contents of an input paper used in the device.

In the drawings, (1) is a paper feed unit, (2) is an input paper, (3) is a character observation unit, (4) is an image memory, and (5)
Is a character extraction unit, (6) is a character buffer, (7) is a recognition unit having a normalization unit (8), a feature detection unit (9) and a collation unit (10), (11) is an output unit, and (12). Denotes a character code memory, (13) denotes a host computer, (14) denotes a keyboard as input means, and (15) denotes a form format storage unit.

Next, the operation of the character recognition apparatus according to the present invention will be described. First, not the cut-sheet paper but the continuous paper is used as the input paper (2), and the individual slip contents (A, B,
C) is continuously described on the input form (2). In this case, the data is printed by an individual data processing device such as a personal computer, and at the time of printing the data A, B, and C, for example, five ＠ marks are provided at the delimiters of the data A, B, and C. Keep it. Then, when the input paper (2) is conveyed to the character observing section (3) through the paper feeding section (1) in the character recognition device, predetermined characters and symbols described on the input paper (2) are sequentially read and the characters are read. A cutout section (5) cuts out a predetermined bit pattern (for example, NXM bits) corresponding to one character from the image memory (4), stores it in a character buffer (6), and adds it to a recognition section (7).

Next, in the recognizing unit (7), a character pattern extracted from the character extracting unit (5) is subjected to recognition processing by a normalizing unit (8), a feature extracting unit (9), and a collating unit (10).

At this time, if a predetermined symbol, that is, five or more “＠” in the above example is entered or printed from the keyboard (14) as an input means, the host computer 1 (13) is set as a break. deep. The data to be entered or printed on the continuous paper can be set in a form associated with various slips or ledgers, and is provided in a form storage unit (15) for storing the form. . Therefore, by the delimiter marks described on the input sheet (2), A and B and B and C are always distinguished regardless of the data content of each data block A, B or C, and the data is read. , Recognition can be done automatically.

When there are a plurality of slip formats or slip formats, a desired format can be selected from the keyboard (14) when the slip format storage unit (15) has a plurality of slip formats.

(G) Effects of the Invention According to the character recognition device of the present invention, the contents of a form such as a voucher based on a predetermined format are described on a continuous sheet, and the data blocks corresponding to the form are automatically separated by a delimiter mark. The operator does not need to perform any operation when recognizing the form.

[Brief description of the drawings]

FIG. 1 is a system block diagram of a character recognition device of the present invention,
FIG. 2 is an explanatory view showing an input sheet used in the apparatus. (2) Input paper, (3) Character observation unit, (5)
... Character extraction unit, (7) recognition unit, (11) output unit, (13) host computer, (14) keyboard, (15) form storage unit.

Claims (1)

(57) [Claims] 1. A character recognition device having a paper feeder for transporting an input sheet, reading a character or a symbol written on the input sheet by a character observing section, applying the output to a recognition section, and recognizing the character. In the apparatus, a character representing a plurality of data based on a predetermined format is described on a continuous paper used as the input paper, and a specific character or a symbol is described in advance as a delimiter mark between the plurality of data, and the character observing unit is described. And a recognition output derived through the recognition unit is added to an output processing unit, the output processing unit detects a break in data block units, and sequentially recognizes characters described in each data block unit. Recognition device.