Hit-Or-Miss 변환을 통한 문자 인식 개선

• 영문명: Character recognition Improvement with Hit-Or-Miss Transformation
• 발행기관: 한국IT마케팅학회
• 저자명: 장성희(Sung Hee Jang) 권영만(Young Man Kwon)
• 간행물 정보: 『한국IT마케팅학회 논문집』2015 IT마케팅학회 논문집 Vol.1 No.1, 149~151쪽, 전체 3쪽
• 주제분류: 경제경영 > 경영학
• 파일형태: PDF
• 발행일자: 2015.11.20

영문 초록

Character recognition in the image have been studied as one of the pattern recognition. In the recent studies, we can find it has been improved significantly in the area of the number or letter recognition. Typically, it is easy to recognize the characters of document with clean characters or good spacing between individual characters. Otherwise, it is hard to do it. In order to resolve this problems, in this paper we proposed the pre-processing algorithm. It is based on a eroded, Hit-Or-Miss and skeletonization operations. As a result, we obtained the better result than existing method.

목차

Abstract

I. 서론

II. 관련연구

III. 제안한 알고리즘과 실험

IV. 결론

참고문헌

키워드

Character recognition Hit-Or-Miss skeletal reduction pretreatment erosion operation expansion operation