ABSTRACT
Online handwriting recognition is a complex research problem that has been addressed by researchers for decades due to its widespread applications. Different methods have been developed by researchers over the years aimed at online handwriting recognition in different Indic scripts texts, and many works are still in progress. One of the most important issues in this field is the availability of standard databases for training, testing and validation of any new or existing algorithm used for the classification. This chapter highlights the challenges faced by researchers in the development of online handwritten character databases and also discusses a way to create corresponding character images to build offline Bangla character databases. This chapter also customizes the existing DHD (directed Hausdorff distance)-based feature extraction technique to generate features from offline character images. The purpose of the work is two-fold: first, to enhance the ability of DHD based feature vector produced from online handwritten characters by incorporating the offline information that is collected from offline character images, and second, to generate character images using online information that helps to build Bangla character image databases that can be used for research in offline mode.
