FUZZY BASED PREPROCESSING USING FUSION OF ONLINE AND OFFLINE TRAIT FOR ONLINE URDU SCRIPT BASED LANGUAGES CHARACTER RECOGNITION

Abstract : Character recognition has been fascinating and intense field of pattern recognition research since early days of computer. This task becomes more challenging when it involves Urdu script based languages especially written in handwritten Nasta’liq font due to the large variations and complexity of the script. Fuzzy logic is an important tool to deal with vague, incomplete, noisy and contradictory information. In order to make handwritten communication with machine more natural, we propose several preprocessing steps for handwritten online Urdu script character recognition to overcome the issue in raw input. We present fuzzy based several preprocessing operations in order to normalize the handwritten stroke using both online and offline domain. The proposed technique is also the necessary step towards character recognition, person identification, personality determination where input data are processed from all perspectives.