Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Computer Science》 2007-09
Add to Favorite Get Latest Update

A New Feature Extraction Method on Handwritten Digits Recognition System

SONG Yue-Cong~(1,2) HU Wei~(1,2) (Department of Computer Science,Mianyang Normal University,Mianyang 621000)~1 (Key Lab.of Program Control and Communication Network,Mianyang Normal University,Mianyang 621000)~2  
This paper discusses the handwritten digits feature extraction,analyzes current three kind of feature extrac- tion method,points out the advantage and weakness,and in this foundation proposed one kind of new feature extrac- tion method,namely 13 points feature extraction method,this method withdrew the key 13 spots as characteristic point from each digits character,it is simply and effective,and saves the feature extraction time,and enhances the recogni- tion system running rate,the experiment indieats,13 points feature extraction method to use in the Handwritten digits recognition have the extremely good compatibility,it has the very big enhancement in the running rate and recognition, tesify the new algorithms efficient and practicable.
【Fund】: 四川省重点科技项目,项目号(02GG006-036)
【CateGory Index】: TP391.4
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved