Abstract | ||
---|---|---|
In this paper we discuss the use of high-level verification on handwritten numeral strings. First of all, we introduce the concept of levels of verification and present the baseline system used to carry out the experiments. Two different strategies were developed: absolute and one-to-one verifiers. A thorough error analysis is also presented in order to identify under which conditions high-level verification is more appropriate. Experimental results are presented on NIST SD19 database. |
Year | DOI | Venue |
---|---|---|
2003 | 10.1016/S0167-8655(02)00226-X | Pattern Recognition Letters |
Keywords | Field | DocType |
different strategy,handwritten digit recognition,neural networks,high-level verification,baseline system,nist sd19 database,handwritten numeral string,one-to-one verifiers,conditions high-level verification,thorough error analysis,numeral string recognition system,neural network | Recognition system,Computer science,Speech recognition,NIST,Baseline system,Artificial neural network,High-level verification,Numeral system | Journal |
Volume | Issue | ISSN |
24 | 7 | Pattern Recognition Letters |
Citations | PageRank | References |
7 | 0.69 | 10 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
L. S. Oliveira | 1 | 385 | 25.17 |
Robert Sabourin | 2 | 908 | 61.89 |
F. Bortolozzi | 3 | 87 | 6.50 |
Ching Y. Suen | 4 | 7569 | 1127.54 |