• V.Kumaresan
• S.Jeya Mary

This study has been undertaken to investigate the Tamil language characters by utilizing various Deep learning methods. RCNN is a sort of Artificial Neural Network (ANN) which is regularly comprises of twofold or triple layers skirts that has nonlinearities and clump standardization. Tamil is one among most conventional dialects particularly found on the southern districts of India. Perceiving a non-digitalized character in Tamil is an extremely challenging cycle in view of its huge and compound person set. In this paper RCNN Architecture which clarifies the RCNN layers and the trouble in acquiring the digitalized character from the RCNN Layer utilizing profound learning. This includes the informational collection preparing and pre-handling of all the Tamil characters. There are absolutely 256 characters in Tamil language in which the vast majority of the letters are practically comparable and just a slight change can be seen distinctly toward the end for a large portion of the characters, so perceiving a specific person is very troublesome and require exceptionally prepared datasets for recognizing every single characters.

Component,formatting,style,styling,insert.

"A Survey on character Recognition based on Deep Learning techniques ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 10, page no.b748-b751, October-2021, Available :http://www.jetir.org/papers/JETIR2110183.pdf

"A Survey on character Recognition based on Deep Learning techniques ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 10, page no. ppb748-b751, October-2021, Available at : http://www.jetir.org/papers/JETIR2110183.pdf