Abstract: A Baseline Based Multi-candidate Mathematical Expression Recognition（BBMMER） method is proposed. Mathematical expression recognition is an important component of pattern recognition, and expression structural analysis is a bottleneck in the development of expression recognition technology. The baseline is used to locate expression-nested structures and analyze the structural relationships between symbols. The recognition results are then expressed in the format of LaTex. The experiments are implemented on a database with a lot of expression images. The results show that the proposed system works well.

Key words: baseline, multi-candidate, mathematical expression recognition, structural analysis

摘要： 提出了一种基于基准线的多候选数学公式识别（Baseline Based Multi-candidate Mathematical Expression Recognition，BBMMER）方法。现代印刷体数学公式识别是模式识别的重要组成部分，而数学公式结构分析又是数学公式识别技术发展的瓶颈所在。提出了一种利用基准线定位公式嵌套结构，多候选分析公式符号间结构关系的方法，并使用LaTex格式表示数学公式的识别结果。在大量的公式图像组成的测试集上取得了良好的公式分析正确率。

关键词: 基准线, 多候选, 数学公式识别, 结构分析