Bit plane decomposition and the scanning n-tuple classifier

S. Hoque1, K. Sirlantzis1, M.C. Fairhurst1
1Department of Electronics, University of Kent, Canterbury, Kent, UK

Tóm tắt

This paper describes a multiple classifier configuration for high performance off-line handwritten character recognition applications. Along with a conventional scanning n-tuple classifier (or sn-tuple) implementation, three other sn-tuple systems have been used which are trained using a binary feature set extracted from the contour chain-codes using a novel decomposition technique. The overall accuracy thus achievable by the proposed scheme is much higher than most other classification systems available and the added complexity (over conventional sn-tuple system) is minimal.

Từ khóa

#Character recognition #Image recognition #Feature extraction #Handwriting recognition #Testing #Electronic mail #Humans #Image segmentation #Fusion power generation #Image coding

Tài liệu tham khảo

sirlantzis, 2001, Genetic algorithm for multiple classifier system configuration: A case study in character recognition, Lecture Notes in Computer Science (LNCS 2096)-Multiple Classifier Systems, 99, 10.1007/3-540-48219-9_10 0, Image processing and computer vision research group, Department of Electronic Engineering schalkoff, 1992, Pattern Recognition Statistical Structural and Neural Approaches 10.1109/TAES.1966.4501787 lucas, 1996, statistical syntactic methods for high-performance ocr, Vision Image and Signal Processing IEE Proceedings-, 143, 23, 10.1049/ip-vis:19960253 10.1049/ic:19960924 10.1145/356625.356627 fairhurst, 1976, A classification system for alpha-numeric characters based on learning network techniques, Digital Processes, 2, 321 10.1049/el:20000494 lucas, 1995, recognition of chain-coded handwritten character images with scanning n-tuple method, Electronics Letters, 31, 2088, 10.1049/el:19951438 10.1109/2.485891 10.1109/ICDAR.2001.953861 hoque, 2000, A moving window classifier for off-line character recognition, Proc Int'l Workshop Frontiers in Handwriting Recognition, 595 10.1049/el:20000494 10.1109/34.667881 10.1007/BF01238023