arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Computer Vision and Pattern Recognition

A Comparative Study of Handwritten Digit Recognition Algorithms

A. Om Surya Prakash, Dilkush Saini, Dhananjay Lakra

See on the map

Abstract

Humans have handwriting styles distinct from human to human. the handwritten text is not always consistent in its size, alignment, or breadth. In order to create a system to comprehend this, it has a computer that can identify and categorise handwritten digit images as ten digits (0 to 9). it is frequently used by academics for a variety of purposes, such as deciphering computerised bank cheque numbers. For this, a variety of deep and machine learning techniques are employed. This study focuses on identifying handwritten digits, namely numbers 0 to 9, using the well known MNIST dataset, which contains 10000 test samples and 60000 training samples. digits are this research presents a comparative review of machine learning technique methods, including convolutional neural network (CNN), decision Tree, logistic regression, and k nearest neighbours (KNN). The dataset is preprocessed and input to the algorithm, and its recall, precision, f1 score, accuracy are determined and compared to test the effectiveness of different algorithms.

More like this