• Priyanka Badhe
• Kalyani Bhalerao
• Arbaj Aga
• S. P. Dhanure
• P.K.Suryawanshi, P.V.Igave

India is a multilingual and multiscript country. Lot of work done on the character recognition of different languages but there is no much work towards the air writing character and multilingual string recognitionAir-writing is the practise of writing characters or words in free space without the use of a hand-held device, utilising finger or hand movements. Air writing varies from traditional handwriting in that the latter includes the pen-up-pen-down action, whereas the former lacks such a clearly defined sequence of occurrences. Due to the simple writing style, it has a significant advantage over the gesture-based system. However, it is a challenging task because of the non-uniform characters and different writing styles. This system proposes a robust and inexpensive system for recognizing multilingual numeral strings in an air writing environment, which uses only a generic device camera for input. The system is divided into two steps; air written multilingual numeral string segmentation and string recognition using deep learning algorithm. This system uses different Indian language for evaluation. CNN based system is used to classify the multilingual Indian numerals.

Air-writing,CNN, gesture-based, human-computer interaction,Vgg16

"CNN BASED NUMERAL RECOGNITION IN AIR WRITING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 8, page no.573-579, August-2022, Available :http://www.jetir.org/papers/JETIRFP06102.pdf

"CNN BASED NUMERAL RECOGNITION IN AIR WRITING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 8, page no. pp573-579, August-2022, Available at : http://www.jetir.org/papers/JETIRFP06102.pdf