Use of Zernike Moments in Handwritten Character Recognition

Abstract

There has been a significant amount of research in pattern recognition in different aspects of handwriting based user interfaces including interactive design tools, ink beautification, and handwritten character recognition. In this thesis, we have focused on the recognition of handwritten Hindi numerals that can be used in common applications like type checking, digital signatures, online document recognition and digital library. Challenges in handwritten characters recognition lie in the variation and distortion of online handwritten numerals since different people may use different style of handwriting, and direction to draw the same shape of any numerals. Handwritten Hindi numerals are imprecise in nature as their corners are not always sharp, lines are not perfectly straight, and curves are not necessarily smooth, unlikely the printed numerals. Furthermore, Hindi numerals can be drawn in different sizes and orientation (the orientation of an arrow depends on its pointing direction), in contrast to handwriting which is often assumed to be written on a baseline in an upright position. Therefore, a robust online handwritten recognition system has to account for all of these factors. In this thesis work, we are considering a statistical approach to Online Handwritten Character Recognition using Zernike moments. We are using Zernike moments as features in the reconstruction of Hindi numerals with its invariance property. Zernike moments have been used in the Optical Character Recognition and Image Recognition applications with good results, but, this feature has not been explored for use in Online Handwritten Character Recognition especially for Hindi numerals recognition.