多字体印刷维吾尔文字符识别系统的研究与开发

该文介绍了维吾尔文的特点及维吾尔文字符识别系统．针对维吾尔文的连体结构．重点讨论了解决过程中的技术难点．其中利用投影分离出连体段中的字母．采用边切分边识别的方法，对文本图像进行了切分．分类．提取外围特征，并通过样张的训练．使维吾尔文字符的识别获得了较满意的结果．

Research and Development of a Multifont Printed Uyghur Character Recognition System

This article introduces the application of the Uyghur sentence characteristic and Uyghur character recognition system to the connected structure of Uyghur. Authors discuss especially those aspects that solve the technological difficult points. The system utilizes projection to isolate the letters of the connected character field. Then they adopt the method of side segmentation and side recognition to perform the segmentation, classification, and extraction of outer features for the Uyghur text image , and the training of the prospectus for Uyghur. Using this character recognition algorithm, we can get relatively satisfying results.