Classification of Tamil Characters using Convolutional Neural Networks

For detailed documentaion can be found at https://www.kaggle.com/dhuruvapriyan/tamil-character-classification/notebook

Dataset

The dataset consists of 3000 images of Tamil Characters from அ to ஓ . Each letter consists of approx. 300 images. Each of images are of different shapes. Dataset can be downloaded from https://drive.google.com/file/d/1fnfdaAAQA-v1l7USrpJAKSWzx3fD0ln8/view?usp=sharing

Data Preprocessing

Extracting images from the dataset and resizing it to shape 224,224. Binarizing the image using opencv-python

Data Classification

Model Definition

Defining Keras Sequential model with CNN's

Training

Prediction