Zoning Methods for Hand-Written Character Recognition: An Overview

Abstract

Zoning is a widespread technique for hand-written character recognition. When a zoning method is considered, the pattern image is subdivided into zones each one providing regional information related to a specific part of the pattern. This paper presents an overview of zoning methods. Through the paper, both static and dynamic zonings are addressed and the most recent approaches for zoning design are discussed, based on genetic algorithms and well-suited zoning representation techniques. Finally, the role of membership functions in zoning-based classification is focused, according to abstract-level, ranked-level and measurement-level weighting models.