印刷体汉字识别技术的研究_毕业设计.doc

PAGE PAGE v 学校代码：10259 上海应用技术学院 学士学位论文 题 目： 硬刷题汉字识别技术的研究 专 业： 电气工程及其自动化 印刷体汉字识别技术的研究 摘要：印刷体汉字识别技术是一种高速、自动的信息录入手段,成为未来计算机的重要职能接口，还可以作为办公自动化、新闻出版、机器翻译等领域的理想输入方式，有着广泛的应用前景。汉字识别的最终目的是使中文信息能更自然，更方便地输入计算机，以便于进一步处理。实际生活中，大量的书信、报纸、杂志内容需要输入计算机，这就是印刷体汉字识别要解决的问题。本文首先在第一章就汉字识别研究的意义和背景以及对目前存在的问题作了综述。第二章对印刷体汉字的识别过程进行基本概述。第三章对印刷体汉字识别作了详细阐述，论述了印刷体汉字识别的原理和预处理、特征抽取与分析、后处理过程；分析了统计模式识别方法、 结构模式识别方法以及人工神经网络模式识别方法；第四章用Matlab对系统进行了仿真；最后在第五章进行全文总结并预测了汉字识别技术研究今后的发展方向。 关键词：汉字识别; 特征提取; 匹配识别; 人工神经网络; Matlab仿真 The research of printed Chinese characters recognition technology Abstract: Printed Chinese character recognition technology is a kind of automatic high-speed, information input method, become the important functions of the computer interface, still can as office automation, the press and publishing, machine translation, etc, the ideal input has wide application prospects. The aim of Chinese character recognition is to make the Chinese input more natural and convenient so that the computer could process Chinese information more easily. In practice, large volume of letters, newspaper, magazines need to be covered into a coded representation of the input characters. Thats what printed Chinese character recognition can do. Firstly, in chapter 1，this thesis gives what is the significance of research and background of Chinese character recognition and introduce the problems we encountered now. The chapter 2，it is a basic overview of the recognition process of printed Chinese characters. In chapter 3, we discusses the principles and printed character recognition, feature extraction and analysis, the post-processing process, Emphatically analyses the statistical pattern recognition method, structural pattern recognition method and Artificial neural network pattern recognition method. I