检索结果-内蒙古大学图书馆

Survey on Handwritten Mathematical Expression recognition in the Last Decade: Grammar- and Graph-Based Parsing, and the Rise of Encoder-Decoder Models and Graph Neural Networks

SSRN

引用

SSRN 2023年

作者： Truong, Thanh-Nghia Nguyen, Cuong Tuan Zanibbi, Richard Mouchère, Harold Nakagawa, Masaki University Research Administrator Center Department of Computer Science Rochester Institute of Technology NY United States Tokyo University of Agriculture and Technology Tokyo Japan FPT University HCMC Campus Viet Nam Document and Pattern Recognition Lab Rochester Institute of Technology New York United States LS2N - UMR CNRS 6004 University of Nantes Nantes France

Machine recognition of handwritten mathematical expressions (HMEs) is an area that has attracted interest owing to steady progress in handwriting recognition and the rapid emergence of pen- and touch-based devices. HME recognition can be considered as an extension of text recognition with a two-dimensional structure of characters and symbols. This survey examines the solutions proposed for recognizing online and offline HMEs over the last decade. A common strategy for recognizing HMEs involves splitting the recognition process into four key tasks: symbol segmentation, symbol classification, spatial relation classification, and structural analysis. Recently, encoder–decoder models using Deep Neural Networks (DNNs) have become popular and can perform all key tasks simultaneously and achieve high performance. Furthermore, evaluation methods and benchmark datasets were used to explore the implicit dependencies among key tasks. Finally, we point out some limitations of the current systems and present a future outlook on the DNN approach and the utilization of context to support HMEs parsing. © 2023, The Authors. All rights reserved.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：