A Projection Based Statistical Approach for Handwritten Character Recognition

Abstract

The aim of this paper is to recognize upper-case English alphabets from handwritten documents. An OCR system based on four-sided projections of an alphabet is proposed. The projection points are approximated by polygons and feature points are extracted subsequently for notch elimination and segmentation of the polygon. The resulting segments are smoothed by Bézier approximation. Statistical matching technique is successfully applied for character recognition.

A Component Recognition-based Chinese Character Segmentation and Structure Discrimination Method

“…They recognized the Chinese character based on feature vector value representing the information of whole image. Pal et al [2] recognized the character by the distance in statistical method. Araki et al [3] proposed a statistical approach for character recognition using Bayesian filter.…”

“…We segment the Chinese character component based on correspondence among contour points and skeleton points. (2) In section III, we analyze the common Chinese character structure and the golden grid construction theory. (3) We give the structure discrimination standards in section IV.…”