计算机工程与应用 ›› 2020, Vol. 56 ›› Issue (18): 1-15.DOI: 10.3778/j.issn.1002-8331.2004-0333

• 热点与综述 • 上一篇    下一篇

场景文字识别技术研究综述

王德青,吾守尔·斯拉木,许苗苗   

  1. 1.新疆大学 信息科学与工程学院,乌鲁木齐 830046
    2.新疆大学 新疆多语种信息技术重点实验室,乌鲁木齐 830046
  • 出版日期:2020-09-15 发布日期:2020-09-10

Review of Research on Scene Text Recognition Technology

WANG Deqing, Wushouer·Silamu, XU Miaomiao   

  1. 1.College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China
    2.Xinjiang Key Laboratory of Multilingual Information Technology, Xinjiang University Urumqi 830046, China
  • Online:2020-09-15 Published:2020-09-10

摘要:

对文字检测和识别技术进行了全面的介绍。介绍了自然场景文字识别技术的研究背景、应用领域、技术难点等;介绍了场景文字识别的预处理技术及流程,介绍了近年来出现的基于深度学习的通用检测网络、维吾尔文和中英文的深度学习文字检测网络、场景文字识别深度学习网络、端到端场景文字检测与识别深度学习网络,并总结了各类网络的结构特点、优势、局限性、应用场景以及实现成本,接着进行了综合分析;最后介绍了公开数据集,并探讨了场景文字识别技术的发展趋势及可能的研究方向。

关键词: 场景文字检测, 文本识别, 深度学习, 端到端检测识别

Abstract:

The text detection and recognition technology is introduced in this paper. Firstly, the research background, application fields and technical difficulties of the natural scene character recognition technology are introduced. Secondly, the preprocessing technology and process of scene text recognition are introduced. Then it introduces the universal deep learning detection network based on deep learning, deep learning text detection network based on Uyghur and Chinese and English, deep learning network based on scene character recognition, end to end scene character detection and recognition deep learning network, this paper summarizes the structure features, advantages, limitations, application scenarios and implementation costs of various networks, and then makes a comprehensive analysis. Finally, it introduces the open data sets, the development trend and possible research direction of scene character recognition technology are also discussed.

Key words: scene text detection, text recognition, deep learning, end-to-end detection and recognition