Deep learning for word-level handwritten Indic script identification

TL;DR

This paper introduces a CNN-based method utilizing multilevel Haar wavelet transforms for feature extraction to identify 11 handwritten Indic scripts, achieving a high accuracy of 94.73%, surpassing existing methods.

Contribution

The paper presents a novel combination of wavelet transforms and multiple CNNs for improved script identification in handwritten Indic scripts.

Findings

Achieved 94.73% script identification accuracy.

Outperformed existing state-of-the-art techniques.

Utilized multilevel Haar wavelet transforms with CNNs for feature extraction.

Abstract

We propose a novel method that uses convolutional neural networks (CNNs) for feature extraction. Not just limited to conventional spatial domain representation, we use multilevel 2D discrete Haar wavelet transform, where image representations are scaled to a variety of different sizes. These are then used to train different CNNs to select features. To be precise, we use 10 different CNNs that select a set of 10240 features, i.e. 1024/CNN. With this, 11 different handwritten scripts are identified, where 1K words per script are used. In our test, we have achieved the maximum script identification rate of 94.73% using multi-layer perceptron (MLP). Our results outperform the state-of-the-art techniques.