A Survey of Sequential Combination of Word Recognizers in Handwritten Phrase Recognition at CEDAR

Authors:
Sargur N. Srihari
Affiliations:
-
Venue:
MCS '00 Proceedings of the First International Workshop on Multiple Classifier Systems
Year:
2000

Citing 4
Cited 0

A theory of multiple classifier systems and its application to visual word recognition

A theory of multiple classifier systems and its application to visual word recognition
A theory of classifier combination: the neural network approach

A theory of classifier combination: the neural network approach
Recognition of offline handwritten words and its extension to phrase recognition

Recognition of offline handwritten words and its extension to phrase recognition
Hierarchical character recognition and its use in handwritten word/phrase recognition

Hierarchical character recognition and its use in handwritten word/phrase recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Several methods for classifier combination have been explored at CEDAR. A sequential method for combining word recognizers in handwritten phrase recognition is revisited. The approach is to take phrase images as concatenations of the constituent words and submit them to multiple recognizers. An improvement of this method takes advantage of the spacing between words in a phrase. We describe the improvements to the overall system as a consequence of the second approach.