Digit Recognition Using Convolution Neural Network

TL;DR

This paper demonstrates that a convolution neural network can achieve over 99% accuracy in digit recognition tasks, improving recognition performance with minimal pre-processing of datasets.

Contribution

The study introduces a CNN-based approach that attains high accuracy in digit recognition without extensive data pre-processing, outperforming traditional machine learning methods.

Findings

Achieved 99.15% accuracy in digit recognition

CNN outperforms KNN, SVM, RFC in this task

Minimal pre-processing required for high accuracy

Abstract

In pattern recognition, digit recognition has always been a very challenging task. This paper aims to extracting a correct feature so that it can achieve better accuracy for recognition of digits. The applications of digit recognition such as in password, bank check process, etc. to recognize the valid user identification. Earlier, several researchers have used various different machine learning algorithms in pattern recognition i.e. KNN, SVM, RFC. The main objective of this work is to obtain highest accuracy 99.15% by using convolution neural network (CNN) to recognize the digit without doing too much pre-processing of dataset.