Devanagari Character Recognition Using MLP-Mixer and CNN Extracted Features

Devanagari is an ancient script that is used to write Hindi, Nepali, Marathi, Maithili, Awadhi, Newari, and Bhojpuri, among other Indo-Aryan languages. Thousands of individuals in India use this script to write documents in Marathi and Hindi. Indian mythology is based on this script. Because of the script's prominence, handwritten Devanagari character identification has grown in popularity over time. In this paper, we present a comparative analysis of different machine learning and deep learning models for Devanagari character recognition. MultiLayer Perceptron(MLP) Mixer is the technology we're using because extensive research has been done in Devanagari Text Recognition using Convolutional Neural Networks(CNN). This paper aims to provide a comparison between traditional models and a newer model(MLP Mixer). We have used CNN as a standalone deep learning model for our problem and used it for feature extraction. We have also developed some traditional machine learning classifiers using these learned complex features.