JPH05274470A - Optical character reader - Google Patents

Abstract

PURPOSE:To obtain the optical character reader which can process a group of documents of different size, received through a communication line, successively. CONSTITUTION:The optical character reader is equipped with at least an FAX adapter 11 which receives document images, a sheet buffer memory 15 which stores the received document images, a sheet information storage part 17 stored with sheet size information corresponding to the IDs of respective documents, and a recognition part 16 which finds the real sizes of the document images by detecting respective guide marks from the stored document images, calculates enlargement/reduction rates from the sheet sizes of the documents corresponding to the IDs and the real sizes of the document images by reading the ID marks, and recognizes the characters of the document images corrected with the enlargement/reduction rates.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical character reading device for reading a form image received by FAX.

[0002]

2. Description of the Related Art Conventionally, an optical character reading device for reading a form image received by a FAX device has been known. Hereinafter, this optical character reading device will be referred to as a FAX-OCR device. In this FAX-OCR apparatus, an input image may expand and contract in the main / sub-scanning direction due to defects in the transport system, optical system, and the like. Therefore, generally, the input image size is compared with the original form size to calculate the image expansion / contraction ratios in the main / sub-scanning directions, and the input image size is corrected based on the expansion / contraction ratio before reading characters. Is going. From this, the conventional FAX-OCR
In the apparatus, the size of the form that can be continuously processed is limited to only one type.

However, with the recent widespread use of FAX-OCR devices, there has been a growing demand for continuous processing of form groups of different sizes, and there is a strong demand for improvement of the device to meet this demand.

[0004]

SUMMARY OF THE INVENTION The present invention has been made in order to meet such a demand, and provides an optical character reading device capable of continuously processing a form group of different sizes received through a communication line. It is an object.

[0005]

In order to achieve the above-mentioned object, the optical character reading device of the present invention has guide marks at at least three corners and has an ID indicating the kind of the form at a predetermined position. An optical character reading device for reading a plurality of types of forms having marks, which stores image data received by the image receiving means for receiving image data of the form transferred through a communication line. Image storage means, form size information storage means in which form size information corresponding to the ID of each form is stored, and each of the guide marks is detected from the form image stored in the image storage means to detect them. The actual size of the form image is calculated based on the position, the ID mark is read, and the form size information corresponding to this ID is obtained. Extracted from the vote size information storage means,
A means for calculating the expansion / contraction ratio of the form image from the extracted form size information and the actual size of the form image, correcting the form image based on this expansion / contraction ratio, and performing character recognition for the corrected form image. It is equipped with.

[0006]

According to the present invention, each guide mark is detected from the form image stored in the image storing means, and the actual size of the form image is obtained based on these detection positions.
The ID mark is read and the form size information corresponding to this ID is extracted from the form size information storage means. Then, the expansion / contraction ratio of the form image is calculated from the extracted form size information and the actual size of the form image, the form image is corrected based on this expansion / contraction ratio, and character recognition is performed on the corrected form image. As a result, it is possible to continuously process form groups of different sizes.

[0007]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a FAX-O according to an embodiment of the present invention.
It is a block diagram for explaining a configuration of a CR device.

In the figure, 11 is a FAX adapter for receiving various data such as image data (compressed data) of a form transferred from an external FAX apparatus through a communication line, and 12 is a FAX controller for controlling the FAX adapter 11. Is. Further, 13 is a CRT / keyboard for displaying and inputting various data such as a form reading result, 1
Reference numeral 4 denotes an image processing unit for performing composite processing of received image data and image processing such as enlarging / reducing and rotating a form image obtained thereby, and 15 denotes an image processing unit 14.
This is a sheet buffer memory for storing the form image that has been composited by. Further, 16 is a recognition unit that performs size check, guide mark search, ID reading, expansion / contraction ratio calculation, etc. on the form image stored in the sheet buffer memory 15 and character recognition, and 17 is a reading target of this apparatus. The sheet information storage unit stores sheet size information, reading control information, and the like corresponding to the IDs of all types of forms. Reference numeral 18 denotes a control unit that performs overall control of the entire device.

FIG. 2 is a diagram showing a form. As shown in the figure, rectangular guide marks 21, 22 and 23 are respectively provided at predetermined positions in the upper left, upper right and lower left corners of this form.
Is printed. Reference numerals 31, 32, and 33 respectively indicate search areas for attempting to search for the guide marks 21, 22, and 23 from all the form images.
Areas 1, 32, and 33 are areas from the upper end of the form to h vertically and from the left end to w horizontally. Reference numeral 41 is an ID mark indicating the type of form, and the position of this ID mark 41 is common to all types of forms, with the upper left end A of the form as the base point.

Next, the operation of this FAX-OCR apparatus will be described.

When the transmission process of the transmitting FAX machine is started, first, the image format, resolution, and form size (form size) of the form image data (compressed data) to be transferred from the transmitting FAX machine to the FAX adapter 11 of this apparatus ( Attribute data such as the horizontal direction) and the transmission rate are transferred, and subsequently, the transfer of the form image data is started.

The FAX adapter 11 sequentially outputs the received form image data to the FAX controller 12. FA
The X controller 12 buffers the form image data in the buffer area in the FAX controller 12 while detecting the line end and the data end in the received form image data. When the buffer is full or when the transmission is completed (when the data end is detected), the form image data in the buffer area is transferred to the control unit 18, and then the attribute data is transferred to the control unit 18.

From the attribute data transferred from the FAX controller 12, the control unit 18 obtains parameters necessary for decoding the image data, such as the decoding method, the number of pixels in the main and sub-scanning directions, and these parameters and The form image data is transferred to the image processing unit 14.

The image processing unit 14 decodes the form image data according to the above parameters, and writes the form image data thus obtained in the sheet buffer memory 15. When the writing of the form image to the sheet buffer memory 15 is completed, the control unit 18 controls the size (height, width) of all types of forms stored in the sheet information storage unit 17.
Image size upper and lower limit values (Wmin-a1, Wmax + a2, Hmin-b1, Hmax + b2;
a1, a2, b1, b2 are preset margins) recognition unit 1
6 is transferred to the sheet buffer memory 1 in the recognition unit 16.
Instruct to check the size of the form image in 5.

The following operation will be described with reference to the flowchart of FIG.

Upon receiving the size check instruction from the control unit 18, the recognition unit 16 first checks the size of the form image stored in the sheet buffer memory 15 in the main scanning direction. This size check is performed on the sheet buffer memory
Size W of the form image stored in 5 in the main scanning direction
If (see FIG. 2) is Wmin −a1 or more and Wmax + a2 or less, it is determined to be valid (step 301).

If the size W in the main scanning direction is valid, the recognition unit 16 next checks the size of the form image stored in the sheet buffer memory 15 in the sub scanning direction.
This size check is judged to be valid if the size H (see FIG. 2) of the form image stored in the sheet buffer memory 15 in the sub-scanning direction is Hmin-b1 or more and Hmax + b2 or less (step 302).

If the size H in the sub-scanning direction is valid, then the recognition unit 16 attempts to search for the guide marks 21, 22, 23 in the order of upper left, upper right, and lower left on the form (step 30).
3-305).

As a result, all the guide marks 21, 2
2, 23 can be searched in the search areas 31, 32, 33,
In addition, if the tilt not exceeding the allowable range of the entire form image based on the positional relationship between the two guide marks in the vertical and horizontal directions is not detected, the recognition unit 16 then determines the guide position based on the detected position of each guide mark (guide mark center position). , The actual size in the main scanning direction (c1 + W1 + c2; where c1 and c2 are fixed values) and the actual size in the sub-scanning direction (d1 + H1 + d2; where d1 and d2 are fixed values) of the form image stored in the sheet buffer memory 15 are obtained. The ID mark 41 is read (step 306).

Next, the recognition unit 16 sends the read ID to the control unit 18. Then, the control unit 18 causes the sheet information storage unit 17 to
From the sheet size (height,
The width) is read out and transferred to the recognition unit 16 (step 307).

After that, the recognition unit 16 compares the received sheet size with the actual size of the form image to calculate the expansion / contraction ratio in the main / sub-scanning direction of the form image stored in the sheet buffer memory 15 ( Step 308). Then, it is checked whether or not the calculated expansion / contraction ratio is within a predetermined allowable range, and the result is notified to the control unit 18 (step 309).

When the control unit 18 receives a notification from the recognition unit 16 that the expansion / contraction ratio is within the allowable range, the control unit 18 instructs the image processing unit 14 to correct the size of the form image based on the expansion / contraction ratio, and the recognition unit 16 also. The read control information corresponding to the ID is transferred to and the reading of characters is instructed. If the form image has an inclination within the allowable range, the image processing unit 14 is instructed to correct the inclination.

As a result, necessary correction is performed on the form image stored in the sheet buffer memory 15 (step 310), after which the recognition unit 16 reads characters for the corrected form image. , And returns the read result to the control unit 18 (step 311). The control unit 18 edits the read result transferred from the recognition unit 16 and outputs the result to the CRT / keyboard 13.

The case where the recognition unit 16 clears all the series of checks on the form image has been described above. However, if any of the checks determines an incorrect image, the control unit 18 regards it as a received image defect and recognizes it. The process of 16 is interrupted there. And CRT / keyboard 1
The contents of the sheet buffer memory 15 are displayed as they are in 3 and the operator is notified of the fact (step 312).

Thus, according to the FAX-OCR apparatus of this embodiment, the guide marks 21, 22 and 23 are detected from the form image stored in the sheet buffer memory 15 and the form image is actually detected based on these detection positions. In addition to obtaining the size, the ID mark 41 is read, the expansion / contraction ratio of the form image is calculated from the sheet size of the form corresponding to this ID and the actual size of the form image, and the form image is corrected based on this expansion / contraction ratio, By performing character recognition on the subsequent form image, it is possible to continuously process form groups of different sizes.

Although the FAX controller 12 and the control unit 18 are separated in this embodiment, the control unit 18 is
The configuration may be such that it replaces the AX controller function. Further, the FAX controller 12 may be a storage type exchange equipped with a reception image buffer.

[0028]

As described above, according to the optical character reading apparatus of the present invention, it is possible to successively process a group of documents of different sizes received through a communication line.

[Brief description of drawings]

FIG. 1 is a block diagram for explaining the configuration of a FAX-OCR device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a form used in the FAX-OCR apparatus of FIG.

3 is a flow chart for explaining the operation of the FAX-OCR apparatus of FIG.

[Explanation of symbols]

 11 ... FAX adapter 12 ... FAX controller 13 ... CRT / keyboard 14 ... Image processing unit 15 ... Sheet buffer memory 16 ... Recognition unit 17 ... Sheet information storage unit 18 ... Control unit

Claims (1)

[Claims] 1. An optical character reading device for reading a plurality of types of forms having guide marks at least at three corners and having ID marks indicating types of forms at predetermined positions, the communication comprising: An image receiving unit for receiving the image data of the form transferred through the line, an image storing unit for storing the form image received by the image receiving unit, and form size information corresponding to the ID of each form are stored. The form size information storage means and the respective guide marks are detected from the form image stored in the image storage means, the actual size of the form image is obtained based on these detection positions, and the ID mark is read to obtain the ID. Form size information corresponding to the form size information is extracted from the form size information storage means, and the extracted form size information and the form A scale factor of the form image is calculated from the actual size of the image, the form image is corrected based on the scale factor, and character recognition is performed on the corrected form image. Optical character reading device.