This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Ninth International Workshop on Frontiers in Handwriting Recognition (IWFHR'04)
Self-Supervised Adaptation for On-Line Text Recognition
Kokubunji, Tokyo, Japan
October 26-October 29
ISBN: 0-7695-2187-8
Loïc Oudot, Laboratoire des Instruments et Systèmes déIle de France
Lionel Prevost, Laboratoire des Instruments et Systèmes déIle de France
Alvaro Moises, Laboratoire des Instruments et Systèmes déIle de France
We recently developed a handwritten text recognizer for on-line text written on a touch-terminal. This system based on the activation-verification cognitive model. It composed of three experts dedicated respectively to signal segmentation in symbols, symbol classification and lexical analysis of the classification results. The baseline system is writer-independent. We present in this paper several strategies of self-supervised writer-adaptation that we compare to the supervised adaptation scheme. The best strategy called "prototype dynamic management" modify the recognizer parameters allowing to get results close to the supervised methods. Results, are presented on a 90 texts (5 400 words) database written by 38 different writers.
Index Terms:
handwriting recognition, self-supervised adaptation, model-based classifier
Citation:
Loïc Oudot, Lionel Prevost, Alvaro Moises, "Self-Supervised Adaptation for On-Line Text Recognition," iwfhr, pp.9-13, Ninth International Workshop on Frontiers in Handwriting Recognition (IWFHR'04), 2004
Usage of this product signifies your acceptance of the Terms of Use.