Feature selection for an automated ancient Tamil script classification system using machine learning techniques

Tamil is one of the oldest languages in the world, spoken in Tamil Nadu, South India, which is inherited from Brahmi Script. The main source of information about history are the stone inscriptions. OCR aids in digitizing Tamil scripts from the ancient and old era to the latest, making its access easy through Internet. Ancient Tamil character recognition from stone inscription is a challenge due to the large disparities of writing style. Efficient feature extraction and selection is essential for effective Ancient Tamil character recognition system. The aim of this paper is the use of Shape and Hough transform for feature extraction using Group Search Optimization and Firefly algorithm for feature selection to recognize the ancient Tamil script.