Robust vision-based features and classification schemes for off-line handwritten digit recognition

DOI10.1016/S0031-3203(01)00228-XzbMath1006.68921MaRDI QIDQ1860209

Kia-Fock Loe, Loo-Nin Teow

Publication date: 20 February 2003

Published in: Pattern Recognition (Search for Journal in Brave)

zbMATH Keywords

feature extractionbiological visionmulticlass classificationlinear discriminationhandwritten digit recognition

Mathematics Subject Classification ID

Pattern recognition, speech recognition (68T10) Computing methodologies and applications (68U99)

Related Items

Unnamed Item, Handwritten digit recognition: Benchmarking of state-of-the-art techniques., A novel cascade ensemble classifier system with a high recognition performance on handwritten digits, A multiple feature/resolution scheme to Arabic (Indian) numerals recognition using hidden Markov models

Uses Software

• SVMlight

Cites Work

• Unnamed Item
• Unnamed Item
• Classification by pairwise coupling
• Support-vector networks
• Statistical Properties of Error Estimators in Performance Assessment of Recognition Systems