A Survey of Methods and Strategies for Feature Extraction in Handwritten Script Identification

Feature extraction is one of the basic function of handwritten script identification. It involves measuring those features of the input pattern are relevant to classification. This paper provides a review of these advances. The aim is to provide an appreciation for the range of techniques that have been developed rather than to simply list sources. Various types of features proposed for handwritten script identification include horizontal and vertical histogram, curvature information and local extreme of curvature, topological features such as loops is a group of white pixels surrounded by black ones, end points is point with exactly 1 neighbouring point, dots a cluster of say 1-3 pixels and junction is a point with more than 2 neighbours all in thinned black and white images, Parameters of polynomial or curve fitting functions, contour information where contours is the outside boundary of a pattern.