Sanskrit Character Recognition using Convolutional Neural Networks: A Survey

Abstract

Sanskrit is language of ancient India with a 3,500-year history and the liturgical language of Hinduism, Buddhism, and Jainism. Current research for Sanskrit Character Recognition from text document images has been one of the difficult due to similarity inside the shapes of different characters, script complexity, non-forte inside the representation and very excessive quantity of symbols.  Sanskrit language is belonging to genre of Devanagari Script. There are many techniques which can be used for recognize characters of scanned document. In this paper we have done survey on Optical Character Recognition (OCR) of devanagari text using long short-term memory (LSTM) and Bidirectional Long-Short Term Memory (BLSTM) neural networks of Sanskrit Characters.