﻿ 基于循环投影统计的数学公式自动定位方法

# 基于循环投影统计的数学公式自动定位方法Mathematical Formula Automatic Location Method Based on Circular Projection Statistics

Abstract: The location of mathematical formulas is the first step to recognize mathematical formula. Only when the formula in the document image is located correctly, one can complete the following steps such as formula symbol recognition, formula document analysis and formula semantic analysis. According to the characteristics of Chinese characters, this paper presents a method for automatic extraction of mathematical formula based on circular projection statistics. This method firstly collects key information through projection, and then extracts the potential line. Finally, mathematical formulas are extracted using a series of constraint conditions. The experimental results show that the method proposed in this work offers correctness of the results at very low computational costs.

[1] 程进. 基本数学公式识别技术的研究[D]. 沈阳工业大学, 2004.

[2] 陈峰, 郑春光. 印刷体文档中的数学公式识别方法综述[J]. 信息技术, 2009, 3: 15-23.

[3] K.-F. Chan, D.-Y. Yeung. Mathematical expression recognition: A survey. International Journal of Oil Document Analysis and Recognition, 2000, 3(1): 3-15.

[4] 丁晓青. 汉字识别研究的回顾[J]. 电子学报, 2002, 30(9): 1364-1368.

[5] 章毓晋. 图象分割[M]. 北京: 科学出版社, 2001.

[6] 刘立波. 图像分割方法探讨[J]. 宁夏农学院学报, 2001, 22(4): 51-56.

[7] 吴冰, 秦志远. 自动确定图像二值化最佳阈值的新方法[J]. 测绘学院报, 2001, 18(4): 283-286.

[8] 张洪刚, 陈光, 郭军. 图像处理与识别[M]. 北京: 北京邮电大学出版社, 2006.

Top