Enhancing Optical Sensor Image Classification through Deep Learning with Convolutional Neural Network

S. Kumarganesh, A. Gopalakrishnan, B. Ragavendran, S. Loganathan, M. Gomathi, I. Rajesh

2025

Abstract

Applications for satellite images are numerous and include environmental surveillance, security, and disaster recovery. Identifying infrastructure and entities in the images by hand is necessary for these applications. Automation is necessary since there are numerous regions to be addressed and a limited number of specialists accessible for performing the searches. However, the issue cannot be resolved by conventional object recognition and categorization techniques since they are too erroneous and unstable. A set of machine learning techniques called deep learning has demonstrated potential for automating these kinds of operations. Through the use of convolutional neural networks, it proved effective at comprehending images. This paper investigates how deep learning can be used to improve the categorization of optical sensor images, with an emphasis on Convolutional Neural Network (CNN). Optical sensor pictures provide insightful information for a variety of uses, including precision farming, tracking the environment, and analysis of land cover. Consequently, to enable more precise and effective categorization, this research makes use of CNN ability to automatically develop hierarchical representations. The effectiveness of deep learning within this field is demonstrated by comparing the model's performance versus conventional classification techniques. The results of this study add to the expanding amount of research in remote sensing utilizing image analysis by offering a solid framework for enhancing the precision and effectiveness of optical sensor picture categorization by utilizing CNN and innovative deep-learning methods. MATLAB is used to implement the suggested framework. The suggested approach outperformed region-based GeneSIS, OBIA, segmentation and classification tree method, fuzzy C means, and segmentation with an accuracy of 95%.

Paper Citation

in Harvard Style

Kumarganesh S., Gopalakrishnan A., Ragavendran B., Loganathan S., Gomathi M. and Rajesh I. (2025). Enhancing Optical Sensor Image Classification through Deep Learning with Convolutional Neural Network. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 676-691. DOI: 10.5220/0013888400004919

in Bibtex Style

@conference{icrdicct`2525,
author={S. Kumarganesh and A. Gopalakrishnan and B. Ragavendran and S. Loganathan and M. Gomathi and I. Rajesh},
title={Enhancing Optical Sensor Image Classification through Deep Learning with Convolutional Neural Network},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={676-691},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013888400004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Enhancing Optical Sensor Image Classification through Deep Learning with Convolutional Neural Network
SN - 978-989-758-777-1
AU - Kumarganesh S.
AU - Gopalakrishnan A.
AU - Ragavendran B.
AU - Loganathan S.
AU - Gomathi M.
AU - Rajesh I.
PY - 2025
SP - 676
EP - 691
DO - 10.5220/0013888400004919
PB - SciTePress