DocumentCode :
317634
Title :
On features used for handwritten character recognition in a neural network environment
Author :
Jameel, Akhtar ; Koutsougeras, Cris
Author_Institution :
Dept. of Comput. Sci. Xavier Univ. of Louisiana, New Orleans, LA, USA
fYear :
1993
fDate :
8-11 Nov 1993
Firstpage :
280
Lastpage :
284
Abstract :
Neural nets are considered as the underlying computing mechanism for a robust approach to the problem of handwritten character recognition. It is expected that recognition mechanisms will be developed through learning algorithms. A key factor to this problem is the set of primitive features which are used to form the raw input vectors representing the digitized image of a character. The authors have explored a number of conventional and new features that can be used in concert with adaptive clustering schemes. Experiences of the performance of these features are presented. A feature which the authors call shadow and which is presented here has produced particularly encouraging results
Keywords :
character recognition; feature extraction; handwriting recognition; neural nets; adaptive clustering schemes; handwritten character recognition; learning algorithms; neural network; primitive features; shadow; Character recognition; Computer networks; Computer science; Decision trees; Handwriting recognition; Information analysis; Intelligent networks; Neural networks; Performance analysis; Robustness;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Tools with Artificial Intelligence, 1993. TAI '93. Proceedings., Fifth International Conference on
Conference_Location :
Boston, MA
ISSN :
1063-6730
Print_ISBN :
0-8186-4200-9
Type :
conf
DOI :
10.1109/TAI.1993.633968
Filename :
633968
Link To Document :
بازگشت