计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (5): 55-58.DOI: 10.3778/j.issn.1002-8331.2010.05.017

• 研发、设计、测试 • 上一篇    下一篇

嵌入式盲人图标实时识别研究

喻正红1,2,石跃祥1,朱珍民2   

  1. 1.湘潭大学 信息工程学院,湖南 湘潭 411105
    2.中国科学院 计算技术研究所,北京 100091
  • 收稿日期:2009-04-01 修回日期:2009-05-19 出版日期:2010-02-11 发布日期:2010-02-11
  • 通讯作者: 喻正红

Blind icon real-time recognition studying on embedded system

YU Zheng-hong1,2,SHI Yue-xiang1,ZHU Zhen-min2   

  1. 1.College of Information Engineering,Xiangtan University,Xiangtan,Hunan 411105,China
    2.Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100091,China
  • Received:2009-04-01 Revised:2009-05-19 Online:2010-02-11 Published:2010-02-11
  • Contact: YU Zheng-hong

摘要: 利用嵌入式系统的特点,提出了一种基于DSP芯片的盲人图标实时识别系统,硬件上构建了系统的基本功能,软件上提出了一种基于不变矩和神经网络的盲人图标识别算法,实验结果表明该系统识别效率较高,识别速度较快,具有携带方便,能源消耗低,可移动性强等特点,满足实际使用的需要,具有一定的应用价值。

关键词: 盲人图标, 模式识别, 不变矩, 神经网络

Abstract: Making full use of features of embedded system,this paper proposes a DSP-based blind icon real-time recognition system.It not only constructs the basic function of the system in hardware but also puts forward an algorithm of blind icon recognition system based on moment invariants and neural network.The experiment shows that it can meet the needs of practical use and has certain application value for its good qualities:High recognition efficiency,quick recognition speed,convenience,low energy consumption,strong mobility and so on.

Key words: blind icon, pattern recognition, moment invariants, neural network

中图分类号: