A NEW APPROACH FOR SLANT ANGLE CORRECTION AND CHARACTER SEGMENTATION OF HANDWRITTEN DOCUMENTS

Abstract

Slant angle correction and segmentation of cursive handwritten letters are two necessary but difficult tasks for any character recognition problem. These two tasks have to be accomplished prior to the recognition of handwritten characters. In this paper we have tried to overcome these difficulties by applying a new approach. The paper is divided into two parts. In the first part, the slant angle of the handwritten characters are determined and removed from a handwritten word by calculating the maximum number of black pixels in any direction and in the next part slant angle free characters are separated from the words depending on the concentration of the pixels which form the character.

Keywords: slant angle, character segmentation, concentration of pixels.