Abstract:
An approach is presented which applies the principle of graph theory and direction information to handwritten numeral recognition. It classifies crudely using graph theory. Primitives are extracted by direction information. The result shows that the approach can get a higher recognition rate.
Key words:
Numeral recognition; Graph theory; Pattern recognition
摘要： 提出一种应用图论的原理和基元方向信息来识别手写数字的方法，根据图论的原理对数字的图像进行粗分类，抽取基元，由方向信息进行细分类，结果表明该方法有比较好的识别结果。
关键词:
数字识别；图论；模式识别
HONG Liurong, WANG Yaocai. Handwritten Numeral Recognition Using Graph Theory and Direction Information[J]. Computer Engineering, 2006, 32(3): 34-36.
洪留荣，王耀才. 应用图论和基元方向信息的手写数字识别[J]. 计算机工程, 2006, 32(3): 34-36.