Avoiding Segmentation in Multi-Digit Numeral String Recognition by Combining Single and Two-Digit Classifiers Trained without Negative Examples

The objective of the present work is to provide an efficient technique for off-line recognition of handwritten numeral strings. It can be used in various applications, like postal code recognition or information extraction from fields of different forms. The proposed solution uses convolutional neur...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Symbolic and Numeric Algorithms for Scientific Computing, Proceedings S. 225 - 230
1. Verfasser: Ciresan, D.
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.09.2008
Schlagworte:
Cellular neural networks
Convolutional codes
convolutional neural network
Handwriting recognition
Image analysis
Image databases
Image recognition
Image segmentation
Neural networks
NIST
NIST SD 19
numeral strings recognition
Pixel
two-digit strings classifier
ISBN:0769535232, 9780769535234
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The objective of the present work is to provide an efficient technique for off-line recognition of handwritten numeral strings. It can be used in various applications, like postal code recognition or information extraction from fields of different forms. The proposed solution uses convolutional neural networks (CNNs) to implement two classifiers, one for digit recognition and one for numeral strings composed from two digits partially overlapped. Both classifiers are trained without negative examples. By comparing the results of the classifiers it can decide if the image contains one digit or two partially overlapped digits. The use of the two-digit strings classifier completely relieves our method from the usage of segmentation. The method is evaluated on a well-known numeral strings database - NIST Special Database 19 - and the results are comparable with the best results from literature, even if those are using elaborate segmentation and training with negative examples.
ISBN:0769535232
9780769535234
DOI:10.1109/SYNASC.2008.68