Digit Recognition by the Implementation of Supervised Learning Using a Convolutional Neural Network
- 1 November 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 227-231
Abstract
The recognition of handwritten numbers is a significant challenge in the field of computer vision, as it finds applications in various domains including banking, medical diagnostics, and handwriting recognition. In recent years, there has been a notable enhancement in our technological capabilities due to the progress made in internet accessibility and technology. Deep learning algorithms have emerged as valuable computational tools for addressing several challenges, including the task of digit recognition. The utilization of convolutional neural networks in the task of digit recognition is driven by their ability to acquire hierarchical representations from picture data. This scholarly article provides a comprehensive examination of the application of Convolutional Neural Networks (CNNs) in the field of machine learning for the purpose of digit recognition. This research examines the architectural components of Convolutional Neural Networks (CNNs), the training procedures employed, and the various methodologies employed for the purpose of digit recognition. The paper also provides an overview of the different datasets used for training and testing CNNs for digit recognition and a discussion of the results obtained from various studies.Keywords
This publication has 13 references indexed in Scilit:
- Early Detection and Classification of Tomato Leaf Disease Using High-Performance Deep Neural NetworkSensors, 2021
- Digit Recognition Using Machine Learning and Convolutional Neural Network2018
- Novel feature extraction technique for the recognition of handwritten digitsApplied Computing and Informatics, 2017
- Simulation of Detecting Function object for AGV Using Computer Vision with Neural NetworkProcedia Computer Science, 2016
- Deep learningNature, 2015
- Handwritten Marathi Character Recognition Using R-HOG FeatureProcedia Computer Science, 2015
- Transfer learning for Latin and Chinese characters with Deep Neural Networks2012
- A novel hybrid CNN–SVM classifier for recognizing handwritten digitsPattern Recognition, 2012
- Best practices for convolutional neural networks applied to visual document analysis2005
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998