JPH0223492A - Optical character reader - Google Patents

Abstract

PURPOSE:To lighten a workload by a human being and to surely set format control information by reading a document for defining physical information and setting the format control information. CONSTITUTION:A document 11 for defining the physical information is optically read through a CCD sensor 15 and stored in a sheet buffer 16. A physical information detecting circuit 19 detects the physical information of the document 11 based on the stored contents, the detected physical information is supplied to an OCR controller 17, and format control (FC) information including the size of the document, a field position in which a character is recorded, a character font, a character type, a pitch, etc., for reading the document is set to the device 17. Further, an OCR reading control circuit 18, etc., are controlled according to the set FC information, and the corresponding document is read. In the above-mentioned setting which is not by a hand, the workload by the human being can be lightened, the FC information can be surely set, and the document can be read accurately.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an optical character reading device that sets format control information by reading a form for defining a reading mode.

(Prior Art) An optical character reader (OCR) optically scans a form to be read, develops a form image obtained on a sheet buffer, and reads the form to be read from this form image. The computer is configured to perform character recognition based on format control information (hereinafter referred to as FC information) corresponding to the FC information. This FC information includes information indicating the size of the form, the position of the field in which characters are recorded, and the character font, character type, pitch, etc. recorded in this field.

For example, when a form to be read is input and the image is developed in the sheet buffer, the type of this form is identified, the characters recorded in the field position indicated by the corresponding FC information are cut out character by character, and the image is expanded into the sheet buffer. Recognize.

The FC information is set in advance before starting the document reading process, but this setting includes the document for reading mode definition (hereinafter referred to as
There is a method using a mode definition sheet (referred to as a mode definition sheet). Information indicating the size of the document to be read, etc. is written in the mode definition sheet.

The OCR reads this mode definition sheet, recognizes each piece of information entered, and stores it as FC information.

By the way, the form size entered in the mode definition sheet,
The physical information of the reading field position is obtained by measuring the actual form by a person using a ruler or the like, and entering the measurement results in the definition sheet.

However, there are usually a plurality of fields to be read, and for each field, it is necessary to measure, for example, the distance from the top of the form, and the starting and ending positions indicating the range of the reading field, which is very time-consuming. Furthermore, since such a plurality of field positions are measured manually, certain errors are likely to occur.

(Problems to be Solved by the Invention) As described above, it is very time-consuming for humans to directly measure the form, and measurement errors are likely to occur. Therefore, if an erroneous measured value is entered in the mode definition sheet as it is and set as FC information, there is a problem that it does not correspond to the form to be read and character recognition cannot be performed.

The present invention has been made in view of the above points, and an object of the present invention is to provide an optical character reading device that can reliably set FC information for character reading processing with a simple operation.

[Configuration of the Invention (Means for Solving the Problems) The present invention stores a form image obtained by scanning a form to be read in an image storage means, and formats the form image stored in the image storage means. In an optical character reading device that performs recognition processing of characters recorded on a reading form based on control information, reading area position information indicating the reading area position targeted for reading processing of the reading form and corresponding to this reading area position information. From the physical information setting form in which reading area designation information is recorded, physical information consisting of the form size and reading area position information of this form and reading area designation information corresponding to this reading area position information are recognized.

Then, the mode definition information recorded in this mode definition form is read out from the mode definition form in which character mode definition information associated with the reading area specification information is recorded. It is configured to recognize characters recorded in area positions as mode definition information, and generate and store format control information used in character recognition processing of reading forms based on this mode definition information and physical information. It is.

(Operation) In the OCR configured in this manner, by inputting a form for physical information definition that has the same shape as the form to be read and has a frame indicating the read field of the form written in, the size of this form can be adjusted. and physical information such as read field position are read from the form and stored as FC information.

(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 OCR according to the same embodiment. In FIG. 1, a form 11 is stably transported to a scanning section (not shown) by an OCR transport mechanism section (not shown). A form 11 shown in the figure is a form 11 for defining physical information, and is used to input physical information of a form to be read. This form 11 has the same shape as the form to be read, and the reading field in which characters are recorded is indicated by a black frame. Furthermore, a field number indicating each field is written in each reading field. The light source 12 generates light for illuminating the form 11 being transported by the transport mechanism. The mirror 13 is the light source 1
The path of the light emitted from 2 is changed to illuminate the document portion located at the reading position (light receiving line). The lens 14 converts the light reflected by the form into a photoelectric conversion unit, such as a CCD sensor 15.
Focus the light upward.

The CCD sensor 15 performs photoelectric conversion on input light according to the amount of light, and is configured by, for example, linearly arranging CODs. This electrical signal from the CCD sensor 15 is binarized in a quantization section (not shown) and stored in the sheet buffer 1B as a form image. The sheet buffer 16 stores photoelectric conversion patterns for one form.

The OCR control device 17 provides processing instructions to the OCR, and provides instructions for reading various forms, FC information for reading forms, and the like. The OCR reading control circuit 18 controls reading processing according to instructions from the OCR control device 17.

When a form image for defining physical information is stored in the sheet buffer 16, the physical information detection circuit 19 detects physical information such as the form size and the position of a reading field indicated by a black frame from this form image.

When the physical information is detected by the physical information detection circuit 19, the character recognition circuit 20 recognizes the field number written in the reading field and outputs it to the physical information detection circuit 19. The physical information detection circuit 19 outputs each piece of information obtained from the physical definition form to the OCR control device 17. O
The CR control device 17 associates each piece of information with information obtained from a mode definition sheet, which will be described later, and controls the FC for reading the form.
Store as information.

Furthermore, when a document image to be read is stored in the sheet buffer 16, the character recognition circuit 20 recognizes the characters recorded in the document image. That is, based on the FC information given from the OCR control device 17, a character pattern for one character is cut out from the form image stored in the sheet buffer 16, and character patterns are extracted by comparing each character with a reference character pattern. Recognize.

Next, the operation of this embodiment will be explained.

The OCR in this embodiment allows the OCR to read the physical information definition form 11 and the mode definition sheet 21 corresponding to this form in advance before starting the form reading process. Set information.

First, the physical information definition form 11 shown in FIG. 2 will be explained. This form 11 is the original paper of the form to be read by the OCR, that is, it is a form on which no characters or the like to be subjected to the reading process are written. This form 11
, the field portion where the reading process is performed is surrounded by a black frame. Furthermore, within the frame of each reading field, a field number indicating this field is written. This field number is set and filled in so as not to be duplicated.

Next, the mode definition sheet 21 shown in FIG. 3 will be explained. This definition sheet 21 corresponds to the form 11 for defining physical/physical information, and information about the characters recorded in each reading field of the form to be read is entered. In other words, the number entered in the field number column corresponds to the field number of the reading field of the form 11, and each column includes the type of font (font) of the characters written in the reading field, kana characters, numbers, Information indicating character types (subsets) such as alphabetic characters, pitch (pitch) indicating character spacing, etc. is entered for each.

The FC information is set using the physical definition form 11 configured as described above and the mode definition sheet 21. First, the form 11 is input into the transport mechanism section of the OCR. Form 1
1 reaches the light receiving line on the conveying surface, the form image of this portion is stored in the sheet buffer 1B. Therefore, when the light receiving line is passed from the front end to the rear end of the form 11, a form image for one form is stored. When the form image of the form 11 is stored in the sheet buffer 16, the physical information detection circuit 19 scans the sheet buffer 1B,
Detect the form size and the position of the black frame indicating the reading field on the form. As for the document size, the vertical size L and the horizontal size W shown in FIG. 2 are detected. Also, the position of the black frame is the distance P1 to P4 from the top edge of the form to the center of the frame, and the distance MSI from the left edge of the form to the left edge of the frame for each field.
~MS4 and the distance from the left edge of the form to the right edge of the frame MEI~ME
4 (that is, the positions of the start and end points indicating the range of each field). Furthermore, the character recognition circuit 20 performs recognition processing on the characters (numbers) written within the black frame from the form image stored in the sheet buffer 16.

The character recognition circuit 20 converts the numbers obtained as a result of this recognition into
It is output to the physical information detection circuit 19 as a field number indicating each read field. In this manner, after detecting each piece of physical information on the physical information definition form 11, the physical information detection circuit 19 transfers the form size information W to the OCR control device 17. Subsequently, position information P1 to P4 for each reading field is provided. MSI~MS4. Transfer ME1 to ME4. The OCR control device 17 stores the physical information transferred from the physical information detection circuit 19. OCR control device 1
7, the stored physical information can be modified by an operator's instructions.

Next, a mode definition sheet 21 for setting the reading mode for each reading field is input to the transport mechanism section of the OCR. When the photoelectric variation m pattern of the mode definition sheet 21 is stored in the sheet buffer 16, the form identification information (ID) is recognized according to instructions from the OCR control device 17. The OCR control device 17 provides the OCR with FC information according to the recognized ID information (roolJ in FIG. 3).

The character recognition circuit 20 performs character recognition processing on the image stored in the sheet buffer 16 based on the FC information for the definition sheet 21 . The character information obtained as a result, that is, information for setting the mode of the reading field, is output to the OCR control device 17. The OCR control device 17 associates the mode definition information recognized from the mode definition sheet 21 with the field number obtained from the physical information definition form 11. In other words, information such as the font, subset, etc. corresponding to the field number "01" recorded on the mode definition sheet 21 is
It is associated as mode definition information for the reading field indicated by field number "1". The OCR control device 17 stores the information obtained in this way as FC information for recognizing characters recorded on the form to be read.

After the FC information is set in this way, the form to be read is input to the transport mechanism section of the OCR. In this form, a frame indicating a reading field is shown in a dropout color, and characters are recorded within this frame. When the form image obtained from this form is stored in the sheet buffer 1B, the FC information set by the form 11 for physical information definition and the mode definition sheet 21 is given to the OCR from the OCR control device. The character recognition circuit 20 is F
Character recognition processing is performed on the form image stored in the sheet buffer 16 based on the C information.

In this way, F used when performing character recognition processing.
The form size and reading field position of the C information can be set by directly reading an actual form in which a frame indicating the field position is recorded. This eliminates the need for the work of measuring forms using a ruler or the like, which has traditionally been done by humans, so that work efficiency can be improved and FC information can be set reliably.

[Effects of the Invention] As described above, according to the present invention, by directly reading a form for physical information definition and setting the form size and reading field position of FC information used for form reading processing,
It is possible to reduce the burden of work previously performed by humans and to reliably set FC information. This makes it possible to accurately perform character recognition processing.

[Brief Description of 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, and Fig. 2 shows the structure of a form for defining physical information used in the embodiment. The figure shown in FIG. 3 is a diagram showing the structure of a form for defining a reading mode used in the same embodiment. 16... Sheet buffer, 17... OCR control device, 18... OCR reading control circuit, 19...
Physical information detection circuit, 20...Character recognition circuit.

Claims (1)

[Claims] A form image obtained by scanning a form to be read is stored in an image storage means, and the form image stored in the image storage means is recorded on the read form based on format control information. In an optical character reading device that performs character recognition processing, a physical device in which reading area position information indicating the reading area position to be subjected to reading processing of the reading form and reading area designation information corresponding to this reading area position information is recorded. a form for setting information, a physical information detection means for detecting physical information consisting of the form size of the physical information setting form and the above-mentioned reading area position information, and the above-mentioned reading area position information detected by this physical information detection means. a reading area designation information recognition means for recognizing the corresponding reading area designation information; a mode definition form in which character mode definition information is recorded, which is associated with the reading area designation information; recognition means for recognizing the mode definition information recorded in the form of the document as the mode definition information of the characters recorded in the reading area position indicated by the reading area designation information; and the mode definition information recognized by the recognition means and the physical An optical character reading device comprising: generating means for generating and storing format control information used in character recognition processing of the reading form based on the physical information detected by the information detecting means. .