On-Line and Off-Line Handwriting Recognition: A Comprehensive Survey
IEEE Transactions on Pattern Analysis and Machine Intelligence
Writer Adaptation for Online Handwriting Recognition
Proceedings of the 23rd DAGM-Symposium on Pattern Recognition
Comparison of Confidence Measures for Face Recognition
FG '00 Proceedings of the Fourth IEEE International Conference on Automatic Face and Gesture Recognition 2000
Comparing Normalization and Adaptation Techniques for On-Line Handwriting Recognition
ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 3 - Volume 3
| Hi-index | 0.00 |
In this paper a writer-independent on-line handwriting recognition system is described comparing the effectiveness of several confidence measures. Our recognition system for single German words is based on Hidden Markov Models (HMMs) using a dictionary. We compare the ratio of rejected words to misrecognized words using four different confidence measures: One depends on the frame-normalized likelihood, the second on a garbage model, the third on a two-best list and the fourth on an unconstrained character recognition. The rating of recognition results is necessary for an unsupervised retraining or adaptation of recognition systems as well as for a user friendly human-computer interaction avoiding excessive call backs.