Hindi-Character-Recognition

This is a Character Recognition System to recognize Devanagari Script. The learning model was trained on 70 thousand images (32x32 pixels) of 64 characters, digits 0 to 9. The optimal result of 97.2% accuracy was achieved.This code helps classify different alphabets of hindi language (Devanagiri) using CNN.

Convolutional Neural Networks

A convolutional neural network (CNN, or ConvNet) is a class of deep neural networks, most commonly applied to analyze visual imagery.

Required Modules

1. numpy
2. pandas
3. opencv
4. tensorflow
5. Keras

Uploaded Files Information

1. data.csv - dataset which is further divided into training-set and test-set
2. devanagari.h5 - Building model
3. Model.py - Training module
4. devanagari_Handwriting_Recognition - Prediction module