Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{169089,
        author = {Kamal dev kumar and Prabhav Mishra and Kanika and Dhananjay kumar and Harmanpreet singh and Deeksha sonal},
        title = {Hand written digit recognition system using deep learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2024},
        volume = {11},
        number = {6},
        pages = {493-498},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=169089},
        abstract = {Identification of human written digits is vital for services like bank check processing, form or document digitalization, and other fields related to computer vision and understanding structure or pattern. Due to the wide variation in handwriting styles, traditional machine learning techniques that rely on manually created characteristics have difficulty. Issues related to the identification of images by machine have altered in recent years by allowing automatic feature extraction from raw data by Convolutional Neural Networks (CNN), which is part of deep learning. In this research, we trained and evaluated a deep learning system for understanding digits written by a person using the MINIST dataset. By using a Keras framework, a feed-forward neural network was designed by the system. A layer for input, two hidden layers with the activation function of the rectified linear unit (ReLU), and an output layer with the softmax function make up the structure. After the equation was optimized with the Adam optimizer and trained using the categorical cross-entropy loss function, it was adjusted. This paper shows how deep learning can generalize across different handwriting styles with excellent accuracy on unseen data, all without requiring a lot of feature building. The findings imply that CNNs and other deep learning-based methods perform better on handwritten digit identification tests.},
        keywords = {Features extraction, Keras, feed-forward neural network, Adam optimizer, image classification.},
        month = {November},
        }