Modelizing character allographs in omni-scriptor frame: a new non-supervised clustering algorithm
Pattern Recognition Letters
Static and Dynamic Classifier Fusion for Character Recognition
ICDAR '97 Proceedings of the 4th International Conference on Document Analysis and Recognition
Writer Adaptation of Online Handwriting Models
ICDAR '99 Proceedings of the Fifth International Conference on Document Analysis and Recognition
Constraint Tangent Distance for On-Line Character Recognition
ICPR '96 Proceedings of the International Conference on Pattern Recognition (ICPR '96) Volume IV-Volume 7472 - Volume 7472
Cascade Classifier: Design and Application to Digit Recognition
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Optical character recognition of Gurmukhi script using multiple classifiers
Proceedings of the International Workshop on Multilingual OCR
Off-line handwritten word recognition using multi-stream hidden Markov models
Pattern Recognition Letters
Constructing dynamic frames of discernment in cases of large number of classes
ECSQARU'11 Proceedings of the 11th European conference on Symbolic and quantitative approaches to reasoning with uncertainty
Combining multiple classifiers for faster optical character recognition
DAS'06 Proceedings of the 7th international conference on Document Analysis Systems
| Hi-index | 0.00 |
Handwriting recognition is such a complexclassification problem that it is quite usual now to makeco-operate several classification methods at the pre-processingstage or at the classification stage. In thispaper, we present an original two stages recognizer. Thefirst stage is a model-based classifier that stores anexhaustive set of character models. The second stage is adiscriminative classifier that separates the mostambiguous pairs of classes. This hybrid architecture isbased on the idea that the correct class almostsystematically belongs to the two more relevant classesfound by the first classifier. Experiments on Unipendatabase show a 30% improvement on a 62 classesrecognition problem.