Novel Approach to Oryza Sativa Leaf Disease Detection Using an Xception‑Based Convolutional Neural Network Architecture

B. Vinothkumar, B. Latha, R. Ravichandran, P. Harishraam, M. Kiranraj, V. Rajkumar

2025

Abstract

Aim: This research aims to develop a better Oryza sativa leaf disease detection system with an Xception-based Convolutional Neural Network (CNN) architecture. The approach will increase accuracy and speed in the identification of various rice leaf diseases and correcting the demerits of traditional detection methods. Materials and Methods: There are two groups in the research. Group 1 refers to the ResNet model, a popular deep learning architecture, to identify rice leaf disease. Group 2 refers to the Xception model of depthwise separable convolutions to improve feature extraction and classification accuracy. In this research Xception works better than ResNet with 96% accuracy against ResNet's 92% along with consuming less processing time by 20%. Results: The proposed system showed better accuracy than the ResNet model. The Xception model was sustaining a mean accuracy of 98.36%, while the control ResNet model was sustaining a mean accuracy of 93.67%, which indicates improved performance. The independent samples test showed that it was significant at 0.0001. Conclusion: This research illustrates that the Xception-based model is more accurate and reliable to identify Oryza sativa leaf diseases, resulting in early identification and improved crop management.

Paper Citation

in Harvard Style

Vinothkumar B., Latha B., Ravichandran R., Harishraam P., Kiranraj M. and Rajkumar V. (2025). Novel Approach to Oryza Sativa Leaf Disease Detection Using an Xception‑Based Convolutional Neural Network Architecture. 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 298-303. DOI: 10.5220/0013897000004919

in Bibtex Style

@conference{icrdicct`2525,
author={B. Vinothkumar and B. Latha and R. Ravichandran and P. Harishraam and M. Kiranraj and V. Rajkumar},
title={Novel Approach to Oryza Sativa Leaf Disease Detection Using an Xception‑Based Convolutional Neural Network Architecture},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={298-303},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013897000004919},
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 - Novel Approach to Oryza Sativa Leaf Disease Detection Using an Xception‑Based Convolutional Neural Network Architecture
SN - 978-989-758-777-1
AU - Vinothkumar B.
AU - Latha B.
AU - Ravichandran R.
AU - Harishraam P.
AU - Kiranraj M.
AU - Rajkumar V.
PY - 2025
SP - 298
EP - 303
DO - 10.5220/0013897000004919
PB - SciTePress