• Journal of Digital Convergence
• /
• v.14 no.6
• /
• pp.499-511
• /
• 2016

The aim of this study was to find out effect of the orthographic representation on speech sound segmentation performance. Children's performances of the orthographic representation task and the speech sound segmentation task had positive correlation in words of phoneme-grapheme correspondence and negative correlation in words of phoneme-grapheme non-correspondence. In the case of words of phoneme-grapheme correspondence, there was no difference in performance ability between orthographic representation high level group and low level group, while in the case of words of phoneme-grapheme non-correspondence, the low level group's performance was significantly better than the high level group's. The most frequent errors of both groups were orthographic conversion errors and such errors were significantly more noticeable in the high level group. This study suggests that from the time of learning orthographic knowledge, children utilize orthographic knowledge for the performance of phonological awareness tasks.

• Proceedings of the KIEE Conference
• /
• 1988.07a
• /
• pp.632-635
• /
• 1988

이 논문에서는 필기체 한글 인식에 있어서 입력된 기본자소를 window를 이용한 윤곽선 추적과 삼각분할에 의한 이분점 추출에 의해 각 기본자소가 갖고있는 특징성분을 찾아내고 그 특징성분에 의해 문자의 골격을 추출하여 인식하는 방법을 제안하였다. 윤곽선 추적시 window를 이용함으로 간단한 잡음제거와 추적속도를 증가 시켰으며 삼각분할에 의한 이분점 추출방법을 사용함으로 단순한 윤곽선 추적에 의해 특징성분을 추출하는 방법보다 문자의 특징성분을 정확하게 추출할 수 있다는 장점을 갖는다.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.1
• /
• pp.119-127
• /
• 1998

In this paper, the design of a VLSI-based multilayer neural network is presented, which can be used as a dedicated hardware for character-type segmentation and character-element recogniti on consuming large processing time in conventional software-based Hangul printed-character recognition systems. Also the architecture and its design of a neural coprocessor interfacing the neural network with a host computcr and controlling thc neural network are presented. The architecture, behavior, and performance of the proposed neural coprocessor are justified using VHDL modeling and simulation. Experimental results show the successful rates of character-type segmentation and character-element recognition is competitive to those of software-based Hangul printed-character recognition systems with retaining high-speed.

• Journal of Korea Multimedia Society
• /
• v.8 no.2
• /
• pp.181-191
• /
• 2005

The classification error of 'ㅁ', 'ㅇ' is one of the main causes of incorrect recognition in Korean characters, but there haven't been enough researches to solve this problem. In this paper, a new feature extraction method from Korean grapheme is proposed to recognize 'ㅁ', 'ㅇ'effectively. First, we defined an optimal neighbor-distance selection measure using modified Hausdorff distance, which we determined the optimal neighbor-distance by. And we extracted neighbor-angle feature which was used as the effective feature to classify the two graphemes 'ㅁ', 'ㅇ'. Experimental results show that the proposed feature extraction method worked efficiently with the small number of features and could recognize the untrained patterns better than the conventional methods. It proves that the proposed method has a generality and stability for pattern recognition.

• Proceedings of the KAIS Fall Conference
• /
• 2006.05a
• /
• pp.485-488
• /
• 2006

본 논문은 수평 런 길이와 수직 런 길이를 이용해 필기체 한글 문자의 자획 두께를 구하고, 그 자획 두께를 이용해 입력 문자의 자소를 수평 성분과 수직 성분으로 분리하는 기술을 제안한다. 수평 성분과 수직 성분 분석은 각도와 관계없이 자획 두께와 수평 런 길이의 변화량만을 이용해 구한다. 분리된 수평 성분 자획과 수직 성분 자획은 오프라인 필기체 한글 인식을 위한 요소 기술 중 하나인 자소 분리를 위한 특징이 된다.

• Annual Conference on Human and Language Technology
• /
• 1993.10a
• /
• pp.503-518
• /
• 1993

봉네트는 온라인 한글 필기 글씨 모델이다 [신92]. 글씨를 자소와 연결획의 결합구조로 보고, 각 자소 및 연결획 모델을 정의한 후, 이들을 제자 원리에 따라 네트워크 구조로 설계한 모델이다. 본 논문에서는 봉네트가 소개된 후 지난 일년 동안 수행되었던 실험 및 모델 검증의 결과와 앞으로도 계속될 개선책을 소개하고, 동 모델의 바탕이 된 통계적 인식 이론을 정립하고자 한다.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.4
• /
• pp.1002-1013
• /
• 1996

The segmentation of characters is an important step in the automatic recognition of handwritten text. This paper proposes the segmenting method of off-line handwritten Hangul. The suggested approach is based on the structural characteristics of Hangul. The first step extracts the local features. connected component and strokes from the imput word. In the second step we identify the class of strokes. The third segmenting step specifies WRC(White Run Column) before consonant or horizontal vowel. If the segment is longer than threshold, the system estimates segmenting columns using the consonant-vowel information and column features, and then finds a cornered boundary along the strokes within the estimated segmenting columns.

• Korean Journal of Cognitive Science
• /
• v.11 no.2
• /
• pp.69-78
• /
• 2000

Originally the Interactive Activation Model(IAM) was developed to explain World Superiority Effect(WSE) in the English words. It is known that there is a similar phenomena in Korean characters. In other words people perceive a grapheme better when it is presented as a component of a character than when it is presented alone. We modified the original IAM to explain the Character Superiority Effect(CSE) for Korean characters. However it is also reported that the degree of CSE for Korean characters varies depending on the type of the character. Especially a component between components was reported to be hard to perceive even though it is in a context. It was supposed that this special phenomenon exists for CSE of Korean characters because Korean character is a two-dimensional composition of components(graphemes). And we could explain this phenomenon by introducing weights for the input stimulus which are calculated by taking into account the two-dimensional shape of the character.

• Annual Conference on Human and Language Technology
• /
• 1994.11a
• /
• pp.505-517
• /
• 1994

국어 철자 검색 프로그램 세 개의 검색 능력을 비교하였다. 오류가 없는 파일, 타자시의 전형적인 오류를 포함하는 파일(자소별 오류율 1%), 그리고 광학적 문자인식 프로그램의 전형적인 오류를 포함하는 파일(자소별 오류율 $2.7{\sim}2.9%$) 등에 대하여 한글과 컴퓨터, 한국 마이크로소프트, 핸디 소프트의 워드프로세서에 도구로 포함된 철자검색 프로그램을 수행하였다. 이 세 프로그램 중에서 한글과 컴퓨터의 제품은 정방향 오판율과 오류율 낮은 파일에 대한 역방향 오판율이 낮았고 핸디 소프트의 제품은 오류율이 높은 파일에 대한 역방향 오판율이 낮았다. 세 프로그램 모두 역방향 오판율이 자소별 오류율의 10배 이상이라는 점에 있어서 심각한 문제를 안고 있는 것으로 판단된다.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.530-536
• /
• 2021

Recently, due to the development of deep learning, end-to-end speech recognition, which directly maps graphemes to speech signals, shows good performance. Especially, among the end-to-end models, conformer shows the best performance. However end-to-end models only focuses on the probability of which grapheme will appear at the time. The decoding process uses a greedy search or beam search. This decoding method is easily affected by the final probability output by the model. In addition, the end-to-end models cannot use external pronunciation and language information due to structual problem. Therefore, in this paper conformer with lexicon transducer is proposed. We compare phoneme-based model with lexicon transducer and grapheme-based model with beam search. Test set is consist of words that do not appear in training data. The grapheme-based conformer with beam search shows 3.8 % of CER. The phoneme-based conformer with lexicon transducer shows 3.4 % of CER.