Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (2): 182-184.DOI: 10.3778/j.issn.1002-8331.2011.02.055

• 图形、图像、模式识别 • Previous Articles     Next Articles

Automatic recognition method of chalk handwritten numerals

HU Xiaodong1,LUO Jiancheng1,WU Wei1,CHENG Xi1,SHEN Zhanfeng1,JIA Yinfang2   

  1. 1.Institute of Remote Sensing Applications,Chinese Academy of Sciences,Beijing 100101,China
    2.Baosteel Group Shanghai Meishan Co.,Ltd,Nanjing 210039,China
  • Received:2009-04-27 Revised:2009-06-12 Online:2011-01-11 Published:2011-01-11
  • Contact: HU Xiaodong

手写粉笔数字自动识别方法研究

胡晓东1,骆剑承1,吴 炜1,程 熙1,沈占锋1,贾银芳2   

  1. 1.中国科学院 遥感应用研究所,北京 100101
    2.上海梅山钢铁股份有限公司,南京 210039
  • 通讯作者: 胡晓东

Abstract: Chalk handwriting is widely used in the production or usual life.After analyzing chalk handwriting features,the automatic recognition method of chalk handwritten numerals is proposed.Two main issues of segmentation and recognition are solved through the universal chalk handwritten segmentation algorithm and the hierarchical classification automatic recognition technique based on SVM.The experiment is executed during the application of the automatic recognition system of steel slab numbers,which proves that the method can achieve high recognition ratio.This method can reach the requirements of application,and has high value of application and significance of promotion.

Key words: chalk handwriting extraction, handwritten digit, feature selection, hierarchical classification, automatic recognition

摘要: 手写粉笔字在生产生活中应用广泛,针对手写粉笔字易与背景混淆的特点,提出了手写粉笔数字自动识别方法,分别通过粉笔字普适性分割算法和基于SVM的手写数字分层分类自动识别技术解决了分割和识别两个关键问题,并结合钢材板坯号自动识别系统的应用进行方法的实验,达到了较高的识别率,具有重大的应用价值和推广意义。

关键词: 粉笔字分割, 手写数字, 特征筛选, 分层分类, 自动识别

CLC Number: