DocumentCode :
3142464
Title :
Lazy evaluation for best-first contextual handwriting recognition
Author :
Downton, A.C. ; Lucas, S.M. ; Du, L.
Author_Institution :
Dept. of Electron. Syst. Eng., Essex Univ., Colchester, UK
fYear :
1999
fDate :
20-22 Sep 1999
Firstpage :
589
Lastpage :
592
Abstract :
Lazy evaluation is a best-first state space search method for contextual handwriting recognition which searches an ordered space and applies constraints at the earliest possible opportunity to maximise computational efficiency. Lazy evaluation is well-suited to multi-level hypothesis verification because it operates recursively on a hierarchical semantic tree of context constraints. This paper describes the lazy evaluation algorithm in detail, and proves that, if ordered hypotheses for the lowest-level sub-patterns are provided as inputs, and the combined belief function is monotonic, then overall interpretations are guaranteed to be generated in best-first order. Applications of lazy evaluation to postcode dictionary and address recognition problems are outlined as illustrations of the algorithm
Keywords :
document image processing; handwriting recognition; image recognition; optical character recognition; search problems; state-space methods; address recognition; belief function; best-first contextual handwriting recognition; best-first state space search method; computational efficiency maximisation; constraints; hierarchical semantic context constraint tree; lazy evaluation; lazy evaluation algorithm; lowest-level sub-patterns; multi-level hypothesis verification; ordered space; postcode dictionary; recursive operation; Algorithm design and analysis; Dictionaries; Ear; Electronic switching systems; Handwriting recognition; Identity-based encryption; Optical character recognition software; State-space methods; Systems engineering and theory;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Document Analysis and Recognition, 1999. ICDAR '99. Proceedings of the Fifth International Conference on
Conference_Location :
Bangalore
Print_ISBN :
0-7695-0318-7
Type :
conf
DOI :
10.1109/ICDAR.1999.791856
Filename :
791856
Link To Document :
بازگشت