Rejection Strategies Involving Classifier Combination for Handwriting Recognition

Authors:
Jose A. Rodríguez;Gemma Sánchez;Josep Lladós
Affiliations:
Computer Vision Center Computer Science Department, Edifici O, Campus Bellaterra, 08913 Bellaterra, Spain;Computer Vision Center Computer Science Department, Edifici O, Campus Bellaterra, 08913 Bellaterra, Spain;Computer Vision Center Computer Science Department, Edifici O, Campus Bellaterra, 08913 Bellaterra, Spain
Venue:
IbPRIA '07 Proceedings of the 3rd Iberian conference on Pattern Recognition and Image Analysis, Part II
Year:
2007

Citing 9
Cited 0

Fusion of handwritten word classifiers

Pattern Recognition Letters - Special issue on fuzzy set technology in pattern recognition
On Combining Classifiers

IEEE Transactions on Pattern Analysis and Machine Intelligence
Introducing New Multiple Expert Decision Combination Topologies: A Case Study using Recognition of Handwritten Characters

ICDAR '97 Proceedings of the 4th International Conference on Document Analysis and Recognition
Rejection Strategies and Confidence Measures for a k - NN Classifier in an OCR Task

ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 1 - Volume 1
Confidence Modeling for Verification Post-Processing for Handwriting Recognition

IWFHR '02 Proceedings of the Eighth International Workshop on Frontiers in Handwriting Recognition (IWFHR'02)
Rejection Methods for an Adaptive Committee Classifier

ICDAR '01 Proceedings of the Sixth International Conference on Document Analysis and Recognition
Confidence Measures for an Address Reading System

ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
Rejection Strategies for Offline Handwritten Sentence Recognition

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 2 - Volume 02
Confidence modeling for handwriting recognition: algorithms and applications

International Journal on Document Analysis and Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper introduces a general methodology for detecting and reducing the errors in a handwriting recognition task. The methodology is based on confidence modeling and its main difference is the use of two parallel classifiers for error assessment. The experimental benchmark associated with this approach is described as well as exhaustive results are provided for two real world recognizers on a large database.