한국어 인쇄체 문자 인식을 위한 연결 성분을 이용한 자소와 자소 인식에 관한 연구
A Study on Grapheme and Grapheme Recognition Using Connected Components Grapheme for Machine-Printed Korean Character Recognition
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
Recognition of grapheme is a very important process in the recognition within 'Hangul(Korean written language)' letters using phoneme recognition. It is because the success or failure in the recognition of phoneme greatly affects the recognition of letters. For this reason, it is reported that separation of phonemes is the biggest difficulty in the phoneme recognition study. The current study separates and suggests the new phonemes that used the connective elements that are helpful for dividing phonemes, recommends the features for recognition of such suggested phonemes, databases this, and carried out a set of experiments of recognizing phonemes using the suggested features. The current study used 350 letters in the experiment of phoneme separation and recognition. In this particular kind of letters, there were 1,125 phonemes suggested. In the phoneme separation experiment, the phonemes were divided in the rate of 100%, and the phoneme recognition experiment showed the recognition rate of 98% in recognizing only 14 phonemes into different ones.
Abstract
Recognition of grapheme is a very important process in the recognition within 'Hangul(Korean written language)' letters using phoneme recognition. It is because the success or failure in the recognition of phoneme greatly affects the recognition of letters. For this reason, it is reported that separation of phonemes is the biggest difficulty in the phoneme recognition study. The current study separates and suggests the new phonemes that used the connective elements that are helpful for dividing phonemes, recommends the features for recognition of such suggested phonemes, databases this, and carried out a set of experiments of recognizing phonemes using the suggested features. The current study used 350 letters in the experiment of phoneme separation and recognition. In this particular kind of letters, there were 1,125 phonemes suggested. In the phoneme separation experiment, the phonemes were divided in the rate of 100%, and the phoneme recognition experiment showed the recognition rate of 98% in recognizing only 14 phonemes into different ones.
참고문헌 (12)
- 오현아, "회전 알고리즘을 이용한 자동차 번호판 문자 인식률 향상 및 문자 보안 강화", 보안공학연구논문지, 13(2), pp.83~90 (2016), 보안공학연구지원센터[DOI]
- 진문용, "적응 휴리스틱 분할 알고리즘을 이용한 실시간 차량 번호판 인식 시스템", 정보처리학회논문지. 소프트웨어 및 데이터 공학, 3(9), pp.361~368 (2014), 한국정보처리학회[DOI]
- Kang-San Kim, "Suggestion of Enhanced Korean Character Recognition Technique Using Google Tesseract Open API", Proceeding of Korean Society for Internet Information, 16(1) (2015)
- 정규수, "도로표지 영상에서 IRBP 기반의 문자 영역 추출", 한국ITS학회 논문지, 13(6), pp.1~9 (2014), 한국ITS학회[DOI]
- Kyung-Wha Park, "Handwritten Hangul Recognition using Multi-column Deep Neural Networks", Proceeding of KIPS Spring Conference, 26(1) (2016)
- Min-Soo Kim, "A Study on Implementation of Printed Character Recognition System And Performance Evaluation", Korea Information Processing Society, 7(11), pp.3584~3591 (2000)
- 林吉澤, "문자형식 분류 기반의 인쇄체 문자인식에 관한 연구", 전자공학회논문지 - CI, 40(9), pp.26~39 (2003), 대한전자공학회
- Kil-Taek Lim, "Reestimation of Recognition Result of MLP Classifier for Machine Printed Hangul-Feasibility Study on Softmax Method", Journal of Information & Electronic Technology, 6, pp.93~105 (2007)
- 임길택, "우편물 자동구분기의 구분율 향상을 위한 문자인식기의 구현 방법", 한국산업정보학회논문지, 12(4), pp.15~24 (2007), 한국산업정보학회
- 이득용, "웨이브렛 특징과 순위 기반 인식을 이용한 한글 문서 영상 검색 시스템", 한국콘텐츠학회 논문지, 5(2), pp.229~242 (2005), 한국콘텐츠학회
- Kyong-Ho Lee, "Study on Automation about Painting the Letters to Road Surface", Proceeding of Korea Society of Computer Information(Summer), 24(1), pp.113~116 (2016)
- A. A. Gooch, "Color2gray : Salience-preserving color removal", ACM Transactions on Graphics, 24(3), pp.634~639 (2005)