Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 03010 | |
Number of page(s) | 7 | |
Section | Image Processing and Computer Vision | |
DOI | https://doi.org/10.1051/itmconf/20257003010 | |
Published online | 23 January 2025 |
Advancements and Challenges in Character Recognition: A Comparative Analysis of CNN and Deep Learning Approaches
Graduate School of Arts & Science, Shanghai New York University, 200124 Shanghai, China
Corresponding author: xy2456@nyu.edu
This paper provides a comprehensive review of character recognition technologies, focusing on the application of Convolutional Neural Networks (CNN) and deep learning methodologies. Through an analysis of three key studies, the research highlights the strengths and limitations of current approaches. Study by Zib emphasizes the challenges in segmenting and recognizing English characters using CNN, revealing the need for supplementary techniques to mitigate errors. Research by Nikitha explores the impact of increasing the dimensionality of analysis, demonstrating that higher dimensions improve accuracy but also extend training times. Similarly, work conducted by Pradeep shows that larger vector sizes enhance recognition accuracy but at the cost of greater computational resources. The collective findings suggest that while CNN and deep learning models have significantly advanced character recognition, there remains a need for enhanced segmentation techniques and a balanced approach to optimizing training efficiency and accuracy. Future research should focus on integrating supportive methods to improve segmentation and finding an optimal trade-off between variable complexity and computational efficiency, thereby advancing the practical application of character recognition systems across various domains.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.