http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
필기체 한글 인식을 위한 웨이브렛 변환과 퍼지데이터를 이용한 특징 추출 방법
조한수(HanSoo Cho) 한국정보기술학회 2005 한국정보기술학회논문지 Vol.3 No.4
In this paper, a new feature extraction method for handwritten Hangul recognition is proposed. It transforms the movements of X and Y coordinates in every input stroke into continuous signal waves and extracts global and local features in start and end position of a stroke using wavelets. In order to overcome shape variations of handwritten Hangul characters. these features are transformed into fuzzy data using membership functions. Stroke recognition is perform by feeding fuzzy data as features of the stroke into the neural networks connected in Hangul recognition network. The handwritten character is recognized using the structural properties of Hangul. Finally, experiments for character recognition are performed using the Hangul online database. Experimental results show that the proposed method can achieve a high performance.