KR100235327B1 - Text identification method - Google Patents

Abstract

The present invention relates to an apparatus and method for recognizing a document such as a text editor or an automatic translation system, in which a cutting region is set before and after an i-th point, The present invention relates to a document recognition apparatus having a character separation function and a method thereof.

The method includes dividing an area from document data input through an image input device, separating individual characters, determining a character from the individual character and determining a contact character, Determining if an 'i' point is present; allocating a cut region before and after the 'i' point if there is an 'i' point and separating the contact character accordingly; And recognizing the character.

Description

A method of recognizing documents with "i" dot separator

FIG. 1 is a diagram showing an English contact character. FIG.

FIG. 2 is a block diagram of a conventional document recognition apparatus having an English-language character separation function.

FIG. 3 is a flowchart illustrating an operation of a document recognition method having a conventional English contact character separation function.

FIG. 4 is a flowchart of an operation of a document recognition method having an English contact character separation function using an "i" point according to an embodiment of the present invention.

FIG. 5 is a diagram for explaining a contact character separation result of a document recognition method having an English contact character separation function using an "i" point according to an embodiment of the present invention.

DESCRIPTION OF THE REFERENCE NUMERALS

10: image input unit 20: key input unit

30: control unit 40: monitor

50: Document output section

The present invention relates to a document recognition method having an English contact character separation function using an "i" point, and more particularly, to a document recognition apparatus such as a text editor or an automatic translation system, Quot; i "for separating a contact character by setting a cut region before and after the character.

A word processor, an automatic translation system, and the like require a document recognition device for recognizing characters input through an image input device.

The above-mentioned document recognition apparatus uses a deterministic recognition method and a syntactic recognition method as a document recognition technique. The former method handles the structure and characteristic relationship of a pattern by pattern feature detection, And the latter method selects the element of the pattern as a primitive and recognizes the structure of the pattern by a syntax rule.

Currently, a document recognition apparatus for recognizing English and Korean characters has been practically used. In such a document recognition apparatus, when there is an English contact character as shown in FIG. 1, It is increasing.

Hereinafter, a document recognition apparatus having a conventional English contact character separation function will be described with reference to the accompanying drawings.

FIG. 2 is a block diagram of a document recognition apparatus having a conventional English contact character separation function. As shown in FIG. 2, the conventional document recognizing apparatus having an English-language character separation function includes an image input unit 10 for converting the content of a document into an electrical image signal by using a scanner or a facsimile, A controller 20 for converting the image signal inputted from the image input unit 10 into data according to a command inputted from the key input unit 20 and recognizing the Korean or English characters, A monitor 40 for displaying characters recognized by the control unit 30 on a screen, a document 40 for hard copying characters recognized and edited by the control unit 30 using a printer or a copying machine, And an output unit 50.

The operation of the conventional document recognition apparatus having the English contact character separation function will be described with reference to FIG. When the user inputs a document recognition command from the key input unit 20, the control unit 30 reads the image information on the document inputted from the image input unit 10 constituted by the image device such as a scanner, . In this process, the image input unit 10 converts document data into electrical signals using a photoelectric conversion element such as a CCD (Charge Coupled Device), converts the document data into electrical signals, and converts the digital data into binary data. The control unit 30 performs area division on the document data thus obtained and separates the text data into a character part and an image area. The technique for realizing this is the RLSA (Run-Length Smoothing Algorithm) technique. When the region segmentation is performed, the control unit 30 extracts a character string in the document region, and extracts and separates the individual characters in the document string, and then discriminates whether the characters extracted and separated individually are in Korean or English .

If the individual characters are Hangul characters, the control unit 30 determines whether the characters are contact characters, and if the characters are contact characters, separates them. Next, the control unit 30 separates the ruled-line characters from the separated Hangul characters or individual Hangul characters, and then processes the recognized Hangul characters. However, if the individual characters are not in Korean and are in English, the control unit 30 determines whether or not the characters are contact characters, and separates the contact characters if they are contact characters as shown in FIG. Next, the control unit 30 recognizes the separated English characters or individual English characters. When the Korean or English recognition process is completed, the control unit 30 stores the recognition result in the internal memory, displays the recognition result on the screen of the monitor 40, and then ends the operation.

After editing the recognized character as described above, the user can hard copy the character using the document output unit 50.

However, such a conventional document recognition apparatus having an English-language contact character separation function has a disadvantage in that it searches for a whole area of a character in a vague contact in the process of separating a contact character, There is a problem in that the reliability of the recognition result is deteriorated because the cutting candidate determination area is set by the rough average without considering the "i ". This is because it is very difficult to separate it from contact characters when the letter proportion (character width / letter length) is small, such as "i ", and when several characters are in contact, And it is also necessary to determine in which area the separation should be attempted because it greatly affects the contact character separation and the character recognition rate.

Disclosure of Invention Technical Problem [8] The present invention has been made to solve the above problems and disadvantages of the related art, and it is intended to prevent an error that may occur in the process of separating English contactors, thereby improving the character recognition rate.

Therefore, it is an object of the present invention to provide a method and apparatus for detecting a contact character by dividing an "i" point in a contact character and then setting a cut region around the "i" I ", which is a block after an " i " point, and a block after an " i " point and an " i " point, in order to increase the recognition rate, and a method thereof.

As a means for achieving the object of the present invention described above, a document recognition method having an English contact character separation function includes the steps of: dividing an area from document data input through an image input device, Judging whether or not an i-th point is present in a case where there is an English contact character; and a step of judging whether or not an i-th point is present, A step of allocating a cut-off area and separating the contact character according to the cut-off area, and a step of recognizing the separated character or individual character.

These and other objects and advantages of the present invention will become more apparent from the following detailed description of the embodiments of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, in order to explain the present invention in detail so that those skilled in the art can easily carry out the present invention. .

The configuration of the document recognition device having the English contact character separation function using the "i" point according to the embodiment of the present invention is merely a hardware recognition of the document recognition device having the English English contact character separation function shown in FIG. It is assumed to be used for convenience. The operation of the document recognition apparatus having the English contact character separation function using the "i" point according to the embodiment of the present invention will be described with reference to FIG. When the user inputs a document recognition command from the key input unit 20, the control unit 30 reads the image information on the document inputted from the image input unit 10 constituted by the image device such as a scanner, In this process, the image input unit 10 converts the document data into electrical signals using a photoelectric conversion element such as a CCD, converts the document data into electric signals, converts the digital data into digital data, and outputs the digital data. The control unit 30 performs area division on the document data obtained from the image input unit 10 using a RLSA (Run-Length Smoothing Algorithm) technique or the like. When the region segmentation is performed on the document data, the control unit 30 extracts the character string in the document region, separates the individual characters in the document string, and determines whether the separated character is in Korean or English.

If the separated character is a Hangul character, the control unit 30 determines whether the character is a contact character again. If the separated character is a contact character, the controller 30 separates the contact character, separates the character from the separated Hangul character or individual Hangul character, do.

However, when the separated character is not English, the control unit 30 determines whether the contact character is a contact character. If the separated character is a contact character as shown in FIG. 1, . If there is an 'i' point in the contact character, the control unit 30 virtually separates the front and back of the 'i' point, and if there is no 'i' point, the cutoff area is allocated in that state to separate the English contact character . FIG. 5 is a diagram showing the results of the case where the contact alphabetic characters are separated around the 'i' point. When the contact characters are separated as described above, the control unit 30 recognizes the separated English characters or individual English characters.

When the Korean or English recognition process is completed, the control unit 30 stores the recognition result in the internal memory, displays the recognition result on the screen of the monitor 40, and then ends the operation. After editing the recognized character as described above, the user can output it using the document output unit 50 such as a printer.

As described above, according to the present invention, it is possible to improve the reliability by separating and setting the cutting area to the front and rear area by using the 'i' point in the English letter contacted by the device for recognizing the document such as the document editor or the automatic translation system And a method of recognizing a document having an English contact character separation function using an "i" point for increasing the recognition rate and a method thereof.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention. Therefore, the technical scope of the present invention should not be limited to the contents described in the detailed description of the specification, but should be determined by the claims.

Claims (2)

Dividing an area from document data input through an image input device and separating individual characters; determining a character from the individual character and determining a contact character; A step of allocating a cutting region before and after the point 'i' in the presence of an 'i' point and separating the contact characters according to the cutting region; and a step of recognizing the separated characters or individual characters The method according to claim 1, wherein the step of dividing the document into a plurality of characters comprises the steps of: 2. The method of claim 1, wherein when separating the point 'i', one contact block is divided into three small blocks (block 'i' point + 'i' point + 'i' And a step of dividing the document into a plurality of blocks by using the 'i' point.