A 3D approach for palm leaf character recognition using histogram computation and distance profile features

Abstract

Handwritten character recognition has been a well-known area of research for last five decades. This is an important application of pattern recognition in image processing. Generally 2D scanning is used and the text is captured in the form of an image. In this work instead of regular scanning method, the X, Y co-ordinates are measured using measuroscope at every pixel point. Further a 3D feature, depth of indentation, ‘Z’, which is proportional to the pressure applied by the scriber at that point, is measured using a dial gauge indicator. In the present work the profile based features extracted for palm leaf character recognition are ‘histogram’ and ‘distance’ profiles. The recognition accuracy obtained using the Z-dimension, a 3D feature, is very high and the best result obtained is 92.8 % using histogram profile algorithm.