A benchmark dataset of onlinehandwritten Gurmukhi script words and numerals

Abstract

This paper presents an online handwritten benchmark dataset (OHWR-Gurmukhi) for Gurmukhi script. TIET, Patiala released the unconstrained online handwriting databases, OHWR-GNumerals and OHWR-GScript, which contain isolated strokes samples produced by 190 writers. The OHWR-GNumerals covers 10 stroke classes and OHWR-GScript covers 95 stroke classes to represent the Gurmukhi character set. For data collection, two data sets of Gurmukhi words have been finalized after having a consultation with language experts in order to collect the balanced stroke samples. The preprocessing methods used to prepare these datasets include: size normalization, removing duplicate points, interpolating missing points and re-sampling. The purpose of this benchmark is to create a common platform and make the benchmark dataset publically available for research endeavors in the area of online handwriting recognition.