数字字符识别算法研究 research on numeral character recognition algorithm.pdfVIP

2010年4月 电 子 测 试 Apr.2010 第4期 ELECTRONIC TEST No.4 数字字符识别算法研究 苑全兵 ，黄福 （中北大学 电子与计算机科学技术学院，太原 030051） 摘要：字符识别是模式识别领域内多数课题均会遇到的基本问题。数字字符识别技术是图像处理领域中的一 个重要研究方向，且在各领域的应用越来越广泛。本文描述了一种改进后的数字字符识别处理方案。首先对 输入的数字字符图像进行预处理，并且对二值化后的图像进行梯度锐化和形态学处理以去除图像中背景的噪 声，从数字中选出局部特征，利用结构语句识别的方法进行数字的识别.这个方法提高了识别数字的准确度和 速度，并通过实验证明了其可行性。 关键词：数字识别；梯度锐化；形态学处理；数字轮廓结构特征 中图法分类号:TP391 文献标识码：A Research on numeral character recognition algorithm Yuan Quanbing, Huang Fu (School of Electronics and Computer Science and Technology, North University of China, Taiyuan 030051 China) Abstract: Character recognition is the basic problem of the most projects in the field of Pattern Recognition .The technology of numeral character recognition is an important research direction in image processing area, and the application in various fields is more and more widely. This paper describes a improved treating plan of numeral character recognition. Firstly, the method is to do the pretreatment for the input numeral character image, and by the gradient sharpening and the morphological processing method background noise would be removed from the image binaryzation, and to select local feature from digit, use the method of structure statements recognition to discriminating digit. This method improved the accuracy and speed of the numeral recognition. Experiments has proved its feasibility. Keywords: numeral recognition; gradient sharpening; morphological processing；structural characteristic of numeral outline 0 引言