Skip to main content

Novel Neural Architectures for Recognition of Handwritten Characters

Howells, Gareth, Fairhurst, Michael, Bisset, D.L. (1996) Novel Neural Architectures for Recognition of Handwritten Characters. In: IEE Workshop on Handwriting Analysis and Recognition, 23 May 1996, Savoy Place, London. (doi:10.1049/ic:19960925) (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided)

The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided. (Contact us about this Publication)
Official URL
http://www.dx.doi.org/10.1049/ic:19960925

Abstract

This paper presents an overview of novel networking strategies for neural networks which significantly improves the training and recognition performance of such networks whilst maintaining the generalisation capabilities achieved by existing architectures. A number of different architectures are introduced based on two major principles. The first of these employs RAM-based neurons arranged in multilayer clusters and the second involves modifying the existing weight structure of a back-propagation network to utilise weights taken from a given domain of Clifford algebra. The architectures are described in terms of the structure of the neurons they employ.

Item Type: Conference or workshop item (Paper)
DOI/Identification number: 10.1049/ic:19960925
Uncontrolled keywords: Optical character recognition, Neural network architecture, Artificial intelligence, Multilayer perceptrons, Backpropagation
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800 Electronics (see also: telecommunications) > TK7880 Applications of electronics (inc industrial & domestic) > TK7882.P3 Pattern Recognition
Divisions: Faculties > Sciences > School of Engineering and Digital Arts > Image and Information Engineering
Depositing User: Gareth Howells
Date Deposited: 30 Nov 2016 19:05 UTC
Last Modified: 29 May 2019 18:23 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/59318 (The current URI for this page, for reference purposes)
  • Depositors only (login required):