PCGAN-CHAR: Progressively Trained Classifier Generative Adversarial Networks for Classification of Noisy Handwritten Bangla Characters

TL;DR

This paper introduces PCGAN-CHAR, a progressively trained GAN that classifies noisy handwritten Bangla characters by learning features at multiple resolutions, effectively handling noise without separate denoising steps.

Contribution

The paper presents a novel progressively trained classifier GAN that classifies noisy handwritten characters across multiple datasets without prior denoising.

Findings

Effective classification of noisy characters demonstrated on MNIST, Bangla Numeral, and Basic Character datasets.

Model accurately classifies characters even with high noise levels.

Progressive training improves feature learning at different resolutions.

Abstract

Due to the sparsity of features, noise has proven to be a great inhibitor in the classification of handwritten characters. To combat this, most techniques perform denoising of the data before classification. In this paper, we consolidate the approach by training an all-in-one model that is able to classify even noisy characters. For classification, we progressively train a classifier generative adversarial network on the characters from low to high resolution. We show that by learning the features at each resolution independently a trained model is able to accurately classify characters even in the presence of noise. We experimentally demonstrate the effectiveness of our approach by classifying noisy versions of MNIST, handwritten Bangla Numeral, and Basic Character datasets.