《手写数字识别方法研究》毕业学术论文.doc

毕业设计（论文） 题 目： 手写数字识别方法研究 学 院： 电子与信息工程学院 专 业： 通信工程 班 级： 通信 姓 名： 指导教师： 起讫时间： 手写数字识别方法研究 摘 要 手写数字识别是图像处理和模式识别领域中的研究课题之一。手写数字由于书写者的因素，使其数字图像的随意性很大，如：笔画粗细、字体大小、字体的倾斜度等差异都将直接影响到数字字符的最终识别结果，因此手写数字识别是模式识别领域内最具有挑战性的课题。 本文主要对手写数字识别的方法进行研究。文中描述了手写数字识别的研究背景和发展状况，并且详细介绍了手写数字识别的预处理方法，讨论了灰度化、二值化、去噪、字符分割、归一化、细化、特征提取和分类识别等方法。在此基础上实现了一个数字识别系统。本系统提取数字的粗网格特征，采用BP神经网络进行训练和识别，并且创建了一个图形用户界面对系统的识别效果进行验证。 实验结果表明，本系统其识别正确率达到了预期设想，是行之有效的系统。 关键词：数字识别 粗网格特征 BP神经网络 Recognition of handwritten digits Abstract Handwritten number recognition is one of the research topic field of image processing and pattern recognition. As a result of the written factors that make number images very arbitrary, such as: stroke thickness, font size, the inclination of handwritten numbers, which have a direct impact on the correct identification of numbers. So recognition of handwritten numbers is the most challenging issues of pattern recognition area. The paper aims to study the methods of handwritten number recognition. In this paper, the background and development of the handwritten number recognition method are introduced. Also the preprocessing of the handwritten number recognition is introduced, which discuss the algorithms of binarization, the smooth, character segmentation and refinement. And then it uses the characters of coarse grid to pick up the numbers’ characters. After that it use the BP neural network to train and recognize. At last, a GUI is created to verify the result. The experimental results show that the recognition system’s recognition accuracy meets the expected requirements. The system is effective. ? Key Words: Number recognition; Characters of coarse grid; BP neural network 目 录 TOC \o 1-3 \h \z \u HYPERLINK \l _Toc264113514 摘 要 PAGEREF _Toc264113514 \h I HYPERL