JP3006294B2 - Optical character reader - Google Patents

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, and more particularly to an optical character reading device for reading forms such as slips.

[0002]

2. Description of the Related Art Conventionally, in the use of an optical character reading apparatus for reading a form, the position of each character on the form has to be reduced in order to prevent erroneous reading due to a displacement of a digit in a full amount column or the like or to improve reading performance. In general, a format such as a character type and the like was described in advance. These tasks are tedious and time-consuming.

However, since there is a certain pattern in the character arrangement of each item of the form, when a new form is designed or changed, a similar format is often input.

[0004]

In a conventional form-reading optical character reading apparatus, when the form such as the character type and position of each character on the form is changed by newly creating or changing the form, a new format is required. Must be created. Therefore, one form is required for one type of form, and the number of forms increases each time a form is newly created or changed, and the management method becomes complicated.

Further, when there are many items of the same format in the same form, it is necessary to repeatedly specify the same contents for all the items, and there is a problem that the format specification becomes complicated.

[0006] For reading printed documents and tables,
Although the format is automatically analyzed, it is necessary to correctly determine the content of the description on a normal form
There is a problem that the performance of correctly extracting characters character by character is insufficient at this stage.

An object of the present invention is to solve the above-mentioned problems,
It is an object of the present invention to provide an optical character reading apparatus that can read a form with high operational efficiency by giving information such as a character position and a character type in advance.

[0008]

According to the present invention, there is provided an optical character reading apparatus comprising: an image input unit for optically inputting data on a sheet including a box such as a form to obtain an input image; An enclosing frame detecting unit that detects the enclosing frame, and an enclosing frame identifying unit that identifies the type of the detected enclosing frame,
A description storage unit that stores description contents including character types and position information relating to characters and images in the box corresponding to the type of the box; and the box identified by the box identification unit.
A description extracting unit that reads out and extracts the written content in the enclosing frame corresponding to the type of the frame from the written content storage unit, and performs a reading process by extracting a character or an image in the enclosing frame according to the written content. It is configured to include a recognition unit for performing
Have been .

[0009]

Next, the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of one embodiment of the present invention. The embodiment illustrated in FIG. 1 includes an image input unit 11 that optically captures data on a paper surface such as a form, a frame detection unit 12 that detects a frame included in the input image acquired by the image input unit 11, A box identification unit 13 for identifying the type of the detected box, a description storage unit 14 that stores the contents described in the frame corresponding to the type of the box, and an identification of the box identified by the box identification unit 13. A description content extraction unit 15 for extracting the description content in the surrounding frame from the description content storage unit 14 based on the
And a recognition unit 16 for recognizing the written contents.

Next, the operation of this embodiment will be described.

The data on the form is optically captured by the image input unit 11, and the frame detection unit 12 detects whether a frame is present in the captured input image.

When an enclosing frame is detected, the type of the enclosing frame is determined by the enclosing frame identifying section 13.

When the type of the enclosing frame is determined, the written content extracting unit 15 extracts the written content from the written content storage unit 14 in which the written content relating to the characters and images described in the enclosing frame is stored in advance. The recognition unit 16 extracts characters or images from the enclosed frame on the input image and performs recognition processing by reading the extracted written contents.

FIG. 2 shows an example of a form used in this embodiment. The reading area 22 is surrounded by an enclosing frame 21 on the paper of the form 5, and an identification frame 23 for determining the type of the enclosing frame 21 is provided at the upper left corner of the enclosing frame 21.

In the identification frame 23, the description content storage unit 14
, The box identification code 24 assigned to the box 21 stored therein, for example, "01" is entered.

The description content storage unit 14 stores a box identification code entered in the type box 23 of the box 21 and a character type and a character related to the character and image in the box 21 in which the box identification code 24 is entered. Described content information such as position information is stored.

The frame detection unit 12 detects the position of the frame on the input image acquired by the image input unit 11 and outputs the position information to the frame identification unit 13. The detection of the position of the enclosing frame in the enclosing frame detecting unit 12 can be realized by using a well-known technique, and thus a detailed description is omitted here.

The enclosing frame identifying unit 13 generates an enclosing frame 21 based on the encircling frame position information provided from the enclosing frame detecting unit 12.
Of the box frame identification code 24 described in the box 23 of FIG. For this purpose, first, the identification frame 23 is detected from the surrounding frame 21 at the storage position input from the surrounding frame detecting unit 12. Next, the image data in the identification frame 23 is input to the recognition unit 16, the surrounding frame identification code 24 is read from the image data, and the result is received from the recognition unit 16. The received result is input to the written content extraction unit 15 as the box identification code 24.

In this embodiment, the enclosing frame identification code 24 is represented by a numeral, but may be another type of character or a code such as a bar code. Further, any position may be set within the form.

The description extracting unit 15 includes the box identifying unit 13.
The specified content of the box identification code 24 input from the storage unit 14 is read from the description storage unit 14, and the description information corresponding to the box identification code 24 is input to the recognition unit 16.

As described above, the recognition unit 16 performs a character reading process on the image data in the identification frame 23 extracted by the box identification unit 13, transfers the read result to the box identification unit 13, and extracts the written contents. In accordance with the description contents extracted by the unit 15, reading processing of characters or images in a box on the input image is performed.

[0023]

As described above, according to the present invention, image input
Optically input data on paper such as forms by the department,
The surrounding frame detection unit detects the surrounding frame on the input image and
The frame identification unit identifies the type of the enclosing frame, and the written content extraction unit
The characters and images in the box corresponding to the type of box
A description of the description, including the character type and position information
Extract from the content group stored in advance in the storage unit,
According to the description contents extracted by the recognition unit,
To read the characters or images
According to the previously stored contents corresponding to the
Capable of reading characters and images. For this reason, by automatically detecting the position of the frame on the form and identifying and setting the format required for extracting characters and images in the form, the form can be newly created or changed by creating or changing the form. There is no need to create a format, and it is easy to manage the format. In addition, when there are many items of the same format in the same form, only one item needs to be specified, so that the format can be easily specified.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention.

FIG. 2 is a diagram showing an example of a form used in the embodiment of FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Image input part 12 Surrounding frame detection part 13 Surrounding frame identification part 14 Description content storage part 15 Description contents extraction part 16 Recognition part 21 Surrounding frame 22 Reading area 23 Identification frame 24 Enclosure frame identification code

Claims (1)

(57) [Claims] An image input unit for optically inputting data on a sheet including a box such as a form to obtain an input image; a box detection unit for detecting the box on the input image; An enclosing frame identification unit for identifying the type of the detected enclosing frame; and a description content storage unit for storing a description content including a character type and a position information relating to characters and images in the enclosing frame corresponding to the type of the enclosing frame. Corresponding to the type of the enclosing frame identified by the enclosing frame identifying unit.
A description content extracting unit that reads and extracts the description content in the box from the description content storage unit, and a recognition unit that extracts and reads a character or image in the box according to the description . An optical character reading device comprising: