Downloads: 1 | Views: 140

Research Paper | Computer Science & Engineering | India | Volume 11 Issue 5, May 2022

Feature Selection for Handwriting Digit Recognition Using Convolutional Neural Network

Tarun Patel | Shivansh Tiwari | Vaibhav Dubey | Vaibhav Singh | Dr. Harvendra Kumar

Abstract: Handwriting digit recognition (HDR) and machine learning have both been spurred by Digit Recognition. Both OCR and HDR have their own domain in which they can be used in an HDR system for digit recognition. Numerous alternative strategies have been offered in numerous studies and papers on how to transform text from the paper documented into a machine-readable format. Digit recognition systems may play a critical part in the eventual building of a paperless society by digitally converting and processing the remaining paper documents. Deep learning has newly taken a fundamental turn in the domain of machine learning (ML) credit to the discovery of artificial neural networks (ANN), which has made it more artificially intelligent (AI). Due to its vast variety of applications, deep learning is in use in a variety of fields, including surveillance, health, medicine, sports, robots, and drones. At the heart of deep learning's incredible achievements is the convolutional neural network, which combines ANN with cutting-edge deep learning algorithms. Pattern recognition, phrase classification, voice recognition, face recognition, text classification, document analysis, scene identification, and HDR are just a few of the many uses. The goal of the particular study is to see how the number of hidden layers and epochs affect the correctness of the CNN in recognising handwritten numbers.

Keywords: Handwritten Digit Recognition, HDR, Artificial Neural Network, ANN, Convolutional Neural Network, CNN, MNIST Dataset

Edition: Volume 11 Issue 5, May 2022,

Pages: 1480 - 1485