TY - GEN
T1 - Confidence measures for an address reading system
AU - Brakensiek, Anja
AU - Rottland, Jörg
AU - Rigoll, Gerhard
N1 - Publisher Copyright:
© 2003 IEEE.
PY - 2003
Y1 - 2003
N2 - In this paper the performance of different confidence measures used for an address recognition system are evaluated. The recognition system for cursive handwritten German address words is based on Hidden Markov Models (HMMs). It is essential, that the structure of the address (name, street, city, country) is known, so that a specific small but complete dictionary can be selected. Choosing a wrong dictionary (OOV: out-of-vocabulary) or misrecognize the word, the recognition result should be rejected by means of the confidence measure. This paper points out two aspects: The comparison of four confidence measures for single words - based on the likelihood, a garbage-model, a two-best recognition or a character decoding - And the comparison of using complete or wrong dictionaries. It is shown, that the best confidence measure - The two-best distance - has a quite different behavior using OOV.
AB - In this paper the performance of different confidence measures used for an address recognition system are evaluated. The recognition system for cursive handwritten German address words is based on Hidden Markov Models (HMMs). It is essential, that the structure of the address (name, street, city, country) is known, so that a specific small but complete dictionary can be selected. Choosing a wrong dictionary (OOV: out-of-vocabulary) or misrecognize the word, the recognition result should be rejected by means of the confidence measure. This paper points out two aspects: The comparison of four confidence measures for single words - based on the likelihood, a garbage-model, a two-best recognition or a character decoding - And the comparison of using complete or wrong dictionaries. It is shown, that the best confidence measure - The two-best distance - has a quite different behavior using OOV.
UR - http://www.scopus.com/inward/record.url?scp=84945937803&partnerID=8YFLogxK
U2 - 10.1109/ICDAR.2003.1227676
DO - 10.1109/ICDAR.2003.1227676
M3 - Conference contribution
AN - SCOPUS:84945937803
T3 - Proceedings of the International Conference on Document Analysis and Recognition, ICDAR
SP - 294
EP - 298
BT - Proceedings - 7th International Conference on Document Analysis and Recognition, ICDAR 2003
PB - IEEE Computer Society
T2 - 7th International Conference on Document Analysis and Recognition, ICDAR 2003
Y2 - 3 August 2003 through 6 August 2003
ER -