节点文献
印刷体蒙古文字识别技术中切分方法的设计与实现
A Design and Implementation of Element Segmentationin the Recognition of Printed Mongolian Characters
【摘要】 文字识别是人工智能的一个重要研究领域.有些民族文字识别的成果已转化为商品化软件.蒙文文字识别的研究还处于起步阶段,蒙古文字的结构和书写方式与其他文种相比有许多特点.印刷体蒙文文字识别技术中字元切分的方法,经过实验其切分正确率在98%以上,并已成功地用于印刷体蒙文识别软件中.
【Abstract】 Character recognition is an important research area in Artificial Intelligence.The productions of some ethnological character recognition have been transformed into commercial software,but the research on the recognition of Mongolian characters is now just beginning.There are many characteristics in Mongolian character structure and written pattern in contrast with other kinds of characters.The method of element segmentation in the recognition of printed Mongolian characters is mainly introduced.Experiments show that the accurate segmentation rate of this method is over 98% and the implementation has been successfully employed in the recognition software of printed Mongolian characters.
【Key words】 pattern recognition; Mongolian characters; character recognition;
- 【文献出处】 内蒙古大学学报(自然科学版) ,Acta Scientiarum Naturalium Universitatis Neimongol , 编辑部邮箱 ,2003年03期
- 【分类号】TP391.43
- 【被引频次】21
- 【下载频次】314