Research Challenges in Off-Line Ancient Handwriting Recognition – A Deep Learning Approach

Huge volumes of antient handwriting documents which has a wealth of information and knowledge, in forms of books, manuscripts or scanned images have been existing in various libraries, offices, museums and different archives all over the world. However, in order to further usage of these raw materials, they need to be transformed into a digital form that would allow the users to read, index, brows and query or even to understand further easily. It is a challenge to maintain and understand the paper-based documents. In this paper, some deep learning based approaches are presented to deal with this sort of documents. This paper also proposed a framework of such a system and future directions for the upcoming researchers in the field.