计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (36): 180-183.DOI: 10.3778/j.issn.1002-8331.2010.36.050

• 图形、图像、模式识别 • 上一篇    下一篇

演化硬件在模式识别中的应用综述

王 进,陈 文,康 雄,李丽芳   

  1. 重庆邮电大学 计算机科学与技术研究所,重庆 400065
  • 收稿日期:2010-08-12 修回日期:2010-10-18 出版日期:2010-12-21 发布日期:2010-12-21
  • 通讯作者: 王 进

Survey on applying evolvable hardware for pattern recognition

WANG Jin,CHEN Wen,KANG Xiong,LI Li-fang   

  1. Institute of Computer Science and Technology,Chongqing University of Posts and Telecommunications,Chongqing 400065,China
  • Received:2010-08-12 Revised:2010-10-18 Online:2010-12-21 Published:2010-12-21
  • Contact: WANG Jin

摘要: 演化硬件以其速度快、灵活性强、实时适应等特点,在模式识别应用中易于建立学习时间短、识别速度快、精确分类的高效识别系统。在论述基于演化硬件模式识别技术的体系结构基础上,总结了不同的演化模型和各自的特性,并对各模型适合的应用领域进行了对比分析。介绍了国内外演化硬件模式识别技术研究的主要方向和发展现状,讨论了演化硬件在模式识别应用中的未来发展趋势和亟需解决的问题。

关键词: 演化硬件, 模式识别, 演化算法, 在线演化

Abstract: Evolvable Hardware(EHW) is rapid,flexibile and real-time adaptative,it’s very efficient to build a classification system with the features of short learning time,high processing speed and high classification accuracy.By analyzing the architecture of EHW-based pattern recognition technology,various EHW-based pattern recognition models and their characteristics are summarized.Different application fields of various EHW recognition schemes are presented.The state-of-the-art researches and advantages on applying EHW for pattern recognition applications are discussed.This paper also indicates the challenges and the future research directions of EHW-based pattern recognition technology which should be further studied.

Key words: evolvable hardware, pattern recognition, evolutionary algorithm, online evolution

中图分类号: