• Sanjay Kumar Singh
• Aditya Khamparia

Hand written character recognition is one of task can be related to automate text recognition system. The system can be improved at level of optical character recognition. The authors has used pretrained neural network ResNet-18, ResNet-50 and ResNet-101 for image classification. furthermore result of these pretrained neural networks are combined using majority voting classifier. The minimum training accuracy 74.13% for character number 26, and maximum training accuracy is 100% for character number 33. The minimum testing accuracy 66.67% for character number 7, and maximum testing accuracy is 19.84% for character numbers 21, 31 and 32. The average accuracy for training is 92.775% and average testing accuracy is 91.13% respectively.

Stereo imaging, 3D depth estimation, image processing, object recognition, height estimation.

"Deep Learning Architecture for Large Scale Hand Written Devanagari Character Recognition", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 10, page no.222-228, October 2018, Available :http://www.jetir.org/papers/JETIRDQ06030.pdf

"Deep Learning Architecture for Large Scale Hand Written Devanagari Character Recognition", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 10, page no. pp222-228, October 2018, Available at : http://www.jetir.org/papers/JETIRDQ06030.pdf