标题:
数学公式识别研究现状Research Status of Mathematical Formula Recognition
作者:
刘东明, 陈联, 李明, 张矩
关键字:
数学公式识别, 研究现状, 文档图像, 笔划, 矢量图形Mathematical Formula Recognition, Research Status, Document Images, Strokes, Vector Images
期刊名称:
《Computer Science and Application》, Vol.5 No.6, 2015-06-25
摘要:
文档的编辑和检索要求能够自动识别数学公式，数学公式识别是一个活跃的研究领域，经过多年的发展提出了许多解决方法。公式的输入数据格式有文档图像、笔划、矢量图形、特殊语言等几种形式，不同的输入方式决定数学公式的提取和和识别方式的不同。本文介绍了数学表达式识别邻域的研究现状，讨论了表达的检测、符号识别、结构分析、语义分析等四部分的问题，并提出未来数学表达式的研究方向和热点。In order to search and edit the documents which contain mathematical formulas, we must auto-matically recognize the expression. Mathematical formula recognition is an active research field and many approaches have been proposed over the years. Nowadays, there are several forms of input data format such as document images, strokes, vector images and so on. Different ways of inputs determine the methods to extract mathematical formulas and different ways of mathematical formula recognition. This article describes the currently researching work of mathematical formula recognition, discusses the four components problems in mathematical formula recognition: the detection of expression, symbol recognition, structural analysis, interpretation and so on, and points out the future research directions of mathematical expressions.