KR960011775A - How to separate contact character of character recognition device - Google Patents

How to separate contact character of character recognition device Download PDF

Info

Publication number
KR960011775A
KR960011775A KR1019940024570A KR19940024570A KR960011775A KR 960011775 A KR960011775 A KR 960011775A KR 1019940024570 A KR1019940024570 A KR 1019940024570A KR 19940024570 A KR19940024570 A KR 19940024570A KR 960011775 A KR960011775 A KR 960011775A
Authority
KR
South Korea
Prior art keywords
character
contact
vertical line
information
characters
Prior art date
Application number
KR1019940024570A
Other languages
Korean (ko)
Other versions
KR970004539B1 (en
Inventor
이영태
임종숭
Original Assignee
이헌조
엘지전자 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 이헌조, 엘지전자 주식회사 filed Critical 이헌조
Priority to KR1019940024570A priority Critical patent/KR970004539B1/en
Publication of KR960011775A publication Critical patent/KR960011775A/en
Application granted granted Critical
Publication of KR970004539B1 publication Critical patent/KR970004539B1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/14—Image acquisition
    • G06V30/148—Segmentation of character regions
    • G06V30/153—Segmentation of character regions using recognition of characters or words
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Character Input (AREA)

Abstract

본 발명은 문자인식장치의 접촉문자 분리방법에 관한 것으로서, 이는 다른 문자에 비해 평균폭이 크게 차이가 나는 영문 접촉문자를 분리하는 데 있어서 영문자의 접촉유형에 따라 문자의 요철정보를 이용하여서 정확히 분리하도록 하는 것이다.The present invention relates to a method of separating the contact character of the character recognition device, which is precisely separated by using the uneven information of the character according to the contact type of the alphabet in separating the English contact characters having a significantly different average width compared to other characters. To do that.

이와같은 본 발명의 목적은 스캔과정과 문자열 추출과정을 통해 얻어진 문자열로부터 개별문자를 인식하는 제1문자인식과정과, 상기 개별문자의 인식이 거절된 경우 접촉문자로 판단하여 상기 접촉딘 문자가 일정한 크기의 수직선으로 구성되어 있는지를 판단하는 접촉유형 추출과정과, 상기 수직선이 존재하면 수직선 정보를 이용하여 접촉문자를 분리하고 수직선이 없으면 윤곽선 정보를 이용하여 접촉문자를 분리하는 접촉문자 분리과정과, 상기 분리된 개별문자를 재인식하는 제2문자인식과정으로 이루어짐으로써, 달성된다.The object of the present invention as described above is a first character recognition process for recognizing an individual character from a string obtained through a scanning process and a string extraction process, and if the recognition of the individual character is rejected, the contacted character is determined to be determined as a contact character. A contact type extraction process for determining whether a contact is composed of a vertical line of size, a contact character separation process for separating contact characters using vertical line information if the vertical line exists, and using contact information if there is no vertical line; It is achieved by the second character recognition process of re-recognizing the separated individual character.

Description

문자인식장치의 접촉문자 분리방법How to separate contact character of character recognition device

본 내용은 요부공개 건이므로 전문내용을 수록하지 않았음As this is a public information case, the full text was not included.

제2오는 본 발명을 수행하기 위한 문자인식장치의 구성도.2 is a block diagram of a character recognition device for carrying out the present invention.

제3도는 제2도에 따른 영문인식 신호흐름도.3 is an English signal flow diagram according to FIG.

제4도는 제3도 개별문자 분리과정에서의 접촉문자 분리 신호흐름도.4 is a signal flow of the character separation signal in the character separation process of FIG.

Claims (10)

스캔과정과 문자열 추출과정을 통해 얻어진 문자열로부터 개별문자를 인식하는 제1문자인식과정과, 상기 개별문자의 인식이 거절된 경우 접촉문자로 판단하여 상기 접촉된 문자가 일정한 크기의 수직선으로 구성되어 있는지를 판단하는 접촉유형 추출과정과, 상기 수직선이 존재하면 수직선 정보를 이용하여 접촉문자를 분리하고 수직선이 없으면 윤곽선 정보를 이용하여 접촉문자를 분리하는 접촉문자 분리과정과, 상기 분리된 개별문자를 재인식하는 제2문자인식과정으로 이루어짐을 특징으로 한 문자인식장치의 접촉문자 분리방법.A first character recognition process for recognizing individual characters from a character string obtained through a scanning process and a character string extraction process; and if the recognition of the individual characters is rejected, the first character recognition process is judged as a contact character, and the contact character is composed of vertical lines of a constant size. A contact type extraction process for determining a contact type, a contact character separation process for separating contact characters using vertical line information if the vertical line exists, and a contact character separation process for separating contact characters using contour information if there is no vertical line, and recognizing the separated individual character Method for separating the contact character of the character recognition device, characterized in that consisting of a second character recognition process. 제1항에 있어서, 수직선 정보를 이용하여 접촉문자를 분리하는 과정은 상기 접촉유형 추출과정에서의 수직선이 존재하면 수직선의 갯수 및 위치정보의 특징을 추출하는 단계와, 상기 추출된 수직선의 위치정보를 이용하여 이미 작성된 절단룰에 의해 문자 접촉유형을 분류하는 단계와, 상기 분류된 접촉문자의 수직선 위치 정보에 따라 접촉문자의 절단방향을 결정하는 단계와, 상기 결정된 절단방향에 따라 수직선으로부터의 각 방향으로 평균문자폭 만큼 수직방향의 런렝스 부호화를 구하는 단계와, 상기 구해진 런렝스 부호화에 따라 접촉문자를 절단하는 단계로 이루어짐을 특징으로 한 문자인식장치의 접촉문자 분리방법.The method of claim 1, wherein the process of separating contact characters by using vertical line information comprises: extracting the number of vertical lines and the feature of position information if a vertical line exists in the contact type extraction process; Classifying the character contact type by the cutting rule already prepared using the method, determining the cutting direction of the contact character according to the vertical line position information of the classified contact character, and the angle from the vertical line according to the determined cutting direction. Obtaining a run length coding in the vertical direction by an average character width in a direction; and cutting contact letters according to the obtained run length coding. 제1항에 있어서, 윤곽선 추적을 이용한 분리과정은 상기 수직선 정보가 존재하지 않을경우 접촉문자로부터 요철상(凹凸狀)의 특징을 추출하는 단계와, 상기 추출된 요철상을 이용하여 접촉문자의 절단 범위를 설정하는 단계와, 상기 추출된 요철상의 정보로부터 접촉문자의 유형을 분류하는 단계와, 상기 분류된 접촉문자의유형에 따라 접촉문자의 절단 범위를 재설정하는 단계와, 상기 구해진 요철상의 위치정보로부터 런렝스 부호화를 구하는 단계와, 상기 구해진 런렝스 부호화에 따라 접촉문자를 절단하는 단계로 이루어짐을 특징으로 한 문자인식장치의 접촉문자 분리방법.The method of claim 1, wherein the separating process using contour tracking comprises extracting features of the uneven image from the contact character when the vertical line information does not exist, and cutting the contact character using the extracted uneven image. Setting a range, classifying a type of contact character from the extracted concave-convex information, resetting a cutting range of the contact character according to the classified contact character type, and obtaining the position information of the concave-convex shape. Obtaining a run length encoding from the step; and cutting the contact character according to the obtained run length encoding. 제2항에 있어서, 접촉유형 분류단계는 한 수직선의 좌우에 다른 수직선이 존재하지 않은 경우의 제1룰과 존재하는 경우의 제2룰로 분류함을 특징으로 한 문자인식장치의 접촉문자 분리방법.3. The method of claim 2, wherein the contact type classification step is classified into a first rule when there is no other vertical line on the left and right of one vertical line and a second rule when it exists. 제4항에 있어서, 제1룰은 상기 수직선이 접촉문자 영상의 앞부분, 뒷부분, 중간부분에 존재하는 경우에 따라 좌측방향, 우측방향, 좌우측방향으로 절단하는 룰인 것을 특징으로 한 문자인식장치의 접촉문자 분리방법.The character recognition apparatus of claim 4, wherein the first rule is a rule which cuts the left, right, and left and right directions according to the case in which the vertical line exists in the front part, the rear part, and the middle part of the contact character image. Character Separation Method. 제4항에 있어서, 제2룰은 이웃하는 수직선의 폭이 하나의 문자인 경우, 이웃하는 수직선의 폭이 하나의 문자보다 작은 경우와, 이웃하는 수직선의 폭이 하나의 문자보다 큰 경우에 따라 각각 좌, 우측 방향, 중간부분으로 절단하는 룰인 것을 특징으로 한 문자인식장치의 접촉문자 분리방법.The second rule of claim 4, wherein the width of the neighboring vertical line is one character, the width of the neighboring vertical line is smaller than one character, and the width of the neighboring vertical line is larger than one character. Method for separating the contact character of the character recognition device, characterized in that the rule to cut to the left, right direction, the middle part, respectively. 제6항에 있어서, 이웃하는 수직선의 폭이 하나의 문자보다 큰 경우에는 제1룰을 이용하여 절단하는 것을 특징으로 한 문자 인식장치의 접촉문자분리방법.7. The method of claim 6, wherein when the width of the adjacent vertical line is larger than one character, the character is cut using the first rule. 제3항에 있어서, 접촉문자의 유형분류 단계는 문자의 접촉부분에 따라 상단접촉, 하단접촉 및 접촉유형 정보가 없는 접촉문자로 분리하는 것을 특징으로 한 문자인식장치의 접촉문자 분리방법.The method of claim 3, wherein the type classification step of the contact character is divided into contact characters without upper contact, lower contact, and contact type information according to the contact portion of the character. 제3항 또는 제8항에 있어서, 접촉문자의 분류유형에 따른 절단범위 설정은 상단접촉 및 하단접촉 문자는 요철상의 시작점과 끝점을 찾은 후 이 영역에 존재하는 수직방향의 흑화소 길이를 구하여 흑화수 길이가 가장작은 위치를 접촉문자의 절단부분으로 결정함을 특징으로 한 문자인식장치의 접촉문자 분리방법.The method of claim 3 or 8, wherein the cutting range is set according to the classification type of the contact letter. The upper contact point and the lower contact letter locate the starting point and the end point of the unevenness, and then obtain the length of the vertical black pixel present in this area. A method of separating contact characters of a character recognition device, characterized in that the position of the smallest length is determined as the cut portion of the contact character. 제3항 또는 제8항에 있어서, 접촉문자의 분류유형에 따른 절단범위 설정은 접촉정보가 없는 유형은 절단 할 수 있는 범위를 설정후 런렝스 정보를 이용하여 절단하는 것을 특징으로 한 문자인식장치의 접촉문자 분리방법.The character recognition apparatus of claim 3 or 8, wherein the setting of the cutting range according to the classification type of the contact character is performed by cutting the run length information after setting the cutting range for the type without the contact information. How to separate contact characters from. ※ 참고사항 : 최초출원 내용에 의하여 공개하는 것임.※ Note: The disclosure is based on the initial application.
KR1019940024570A 1994-09-28 1994-09-28 Contacted character separating method of character recognition apparatus KR970004539B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1019940024570A KR970004539B1 (en) 1994-09-28 1994-09-28 Contacted character separating method of character recognition apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019940024570A KR970004539B1 (en) 1994-09-28 1994-09-28 Contacted character separating method of character recognition apparatus

Publications (2)

Publication Number Publication Date
KR960011775A true KR960011775A (en) 1996-04-20
KR970004539B1 KR970004539B1 (en) 1997-03-28

Family

ID=19393794

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1019940024570A KR970004539B1 (en) 1994-09-28 1994-09-28 Contacted character separating method of character recognition apparatus

Country Status (1)

Country Link
KR (1) KR970004539B1 (en)

Also Published As

Publication number Publication date
KR970004539B1 (en) 1997-03-28

Similar Documents

Publication Publication Date Title
Shi et al. Line separation for complex document images using fuzzy runlength
US8059868B2 (en) License plate recognition apparatus, license plate recognition method, and computer-readable storage medium
CN101751567A (en) Quick text recognition method
Lam et al. Reading newspaper text
KR960011775A (en) How to separate contact character of character recognition device
Nguyen et al. Enhanced character segmentation for format-free Japanese text recognition
KR0186172B1 (en) Character recognition apparatus
KR102064974B1 (en) Method for recogniting character based on blob and apparatus using the same
KR950001553A (en) How to Separate Individual Characters in English Strings
KR100480024B1 (en) Collection Recognition Method Using Stroke Thickness Information
JP2728086B2 (en) Character extraction method
KR960002072A (en) Contact Character Separation Method of English Recognition System
JP2570415B2 (en) Character extraction method
JPS6227887A (en) Character type separating system
KR930014166A (en) Individual Character Cutting Method of Document Recognition Device
KR970002740A (en) Contact Character Separation and Feature Extraction Method of Character Recognition Device
JP2993252B2 (en) Homomorphic character discrimination method and apparatus
JP2728085B2 (en) Character extraction method
Moalla et al. Extraction of arabic words from multilingual documents
JP3193573B2 (en) Character recognition device with brackets
JP2520174B2 (en) Automatic character extraction device
JPH09282417A (en) Character recognition device
JPH02230484A (en) Character recognizing device
Garris Teaching Computers to Read Handprinted Paragraphs
Bodduluri et al. A novel way of identifying telugu, tamil and english scripts by priority check using discerning features

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
G160 Decision to publish patent application
E701 Decision to grant or registration of patent right
GRNT Written decision to grant
FPAY Annual fee payment

Payment date: 20050607

Year of fee payment: 9

LAPS Lapse due to unpaid annual fee