Olive Leaf Disease Detection Using Transformer-Based Deep Learning Approach

Shahzada Muhammad Junaid; Rabia Tehseen; Uzma Omer; Muhammad Inam Ul Haq; Ayesha Zaheer

Authors

Shahzada Muhammad Junaid Department of Data Science, University of Central Punjab, Lahore, Pakistan
Rabia Tehseen Department of Data Science, University of Central Punjab, Lahore, Pakistan
Uzma Omer Department of Computer Science, University of Education, Lahore, Pakistan
Muhammad Inam Ul Haq Department of Computer Science, University of Education, Lahore, Pakistan.
Ayesha Zaheer Department of Computer Science and Bioinformatics, Khushal Khan Khattak University, Karak, Pakistan

Keywords:

Artificial Intelligence (AI), Deep Learning, Vision Transformer (ViT), Olive Leaf Disease Detection, Explainable AI, Smart Agriculture

Abstract

The use of AI and DL in automated crop health monitoring and disease diagnosis, especially relevant to Pakistan's burgeoning olive growing industry, has gained momentum. This paper proposes a transformer-based deep learning approach for the detection of olive leaf diseases due to significant shortcomings in the robustness and generalization of traditional convolutional neural networks. The proposed system makes use of a Vision Transformer (ViT) architecture to extract both local and global contextual features from the images of leaves using multi-head self-attention mechanisms. The developed Optimized ViT-Small model identifies olive leaves into three classes: Healthy, Aculus olearius, and Olive Peacock Spot. It is trained and tested on a pre-processed dataset of 3,400 high-resolution olive leaf images collected from olive-growing regions of Pakistan. Experimental results show strong performance with a test accuracy of 97% while demonstrating high precision, recall, and F1-scores throughout the classes. Moreover, performance assessment through confusion matrix analysis, ROC AUC, and precision-recall curves supports the developed model's effectiveness. Although the dataset's geographical coverage is limited, the results indicate that transformer-based architectures are an attractive alternative for the applications of precision agriculture in Pakistan.

References

A. M. ; A.-J. A H ; Alsaeedi, “Dynamic Clustering Strategies Boosting Deep Learning in Olive Leaf Disease Diagnosis,” Sustainability, vol. 15, no. 18, p. 13723, 20231, doi: https://doi.org/10.3390/su151813723.

I. N. C.-C. Erbert F. Osco-Mamani, “Highly Accurate Deep Learning Model for Olive Leaf Disease Classification: A Study in Tacna-Per´u,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 4, 2023, [Online]. Available: https://thesai.org/Publications/ViewPaper?Volume=14&Issue=4&Code=IJACSA&SerialNo=94

J. V. D. Singh, A. Kaur, “Machine vision for agricultural disease identification: a review,” Comput. Electron. Agric., vol. 210, pp. 107–120, 2023.

S. Wang et al., “Advances in Deep Learning Applications for Plant Disease and Pest Detection: A Review,” Remote Sens., vol. 17, no. 4, p. 698, Feb. 2025, doi: 10.3390/RS17040698/S1.

J. L. João Mendes, “Impact of hyper-parameter tuning on CNN accuracy in agricultural image classification,” Smart Agric. Technol., vol. 11, p. 101016, 2025, doi: https://doi.org/10.1016/j.atech.2025.101016.

A. A. Ibrahim Alrashdi, “An efficient deep-learning model for olive tree diseases diagnosis in Al-Jouf region,” Alexandria Eng. J., vol. 130, pp. 709–723, 2025, doi: https://doi.org/10.1016/j.aej.2025.09.059.

M. D. & S. K. Ishak Pacal, Serhat Kilicarslan, Burhanettin Ozdemir, “Efficient and autonomous detection of olive leaf diseases using AI-enhanced MetaFormer,” Artif. Intell. Rev., vol. 58, no. 303, 2025, doi: https://doi.org/10.1007/s10462-025-11131-y.

F. A. Attilio Di Nisio, “Fast Detection of Olive Trees Affected by Xylella Fastidiosa from UAVs Using Multispectral Imaging,” Sensors, vol. 20, no. 17, p. 4915, 2020, doi: 10.3390/s20174915.

H. Alshammari, K. Gasmi, I. Ben Ltaifa, M. Krichen, L. Ben Ammar, and M. A. Mahmood, “Olive Disease Classification Based on Vision Transformer and CNN Models,” Comput. Intell. Neurosci., vol. 2022, 2022, doi: 10.1155/2022/3998193.

Alaoui Fatima-Zahra, Zouheir Banou, Sanaa El Filali, Rachida Ait Abdelouahid, “Deep Learning for Plant Disease Detection: A Novel Convolutional Neural Network Approach Using the new dataset HealthySick Plants,” E3S Web Conf., 2025, doi: 10.1051/e3sconf/202568000005.

F. A.-H. M. Al-Dhaheri, A. Al-Marzooqi, “Multimodal deep learning for olive disease classification using spectral and visual data,” Agriculture, vol. 13, no. 7, pp. 1197–1212, 2023.

A. A. M. Bashir, U. Mehmood, “Vision Transformer for olive leaf disease detection,” Comput. Mater. Contin., vol. 79, no. 2, pp. 2819–2835, 2024.

S. K. H. Hussain, F. Raza, “CNN–Transformer hybrid model for high-accuracy olive leaf disease recognition,” Sensors, vol. 24, no. 8, pp. 3285–3297, 2024.

Z. A. T. Abbasi, M. Khan, “Olive leaf disease detection using convolutional neural networks,” Procedia Comput. Sci., vol. 193, pp. 252–260, 2021.

F. U. A. Uysal, “Convolutional neural network-based feature extraction for plant disease detection,” Expert Syst. Appl., vol. 178, pp. 114–125, 2021.

M. O. A. Mohamed, S. Ali, “MobiRes-Net: A hybrid deep learning architecture for olive leaf disease classification,” Neural Process. Lett., vol. 55, pp. 1543–1561, 20225.

L. L. P. Paredes, R. Alves, “Olive tree disease detection using optimized convolutional models,” Sensors, vol. 24, no. 5, pp. 2310–2321, 2024.

S. T. N. Rana, M. Latif, “Sustainable deep learning approaches for olive disease prediction,” Sustainability, vol. 16, no. 5, pp. 2752–2768, 2024.

U. S. M. Qureshi, H. Ahmad, “Hybrid CNN–LSTM network for automatic olive leaf disease diagnosis,” Neural Comput. Appl., vol. 36, pp. 501–513, 2024.

A. Y. R. Ameen, K. Shakir, “Performance analysis of machine learning algorithms for olive leaf disease classification,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 3, pp. 54–61, 2023.

F. J. T. Hamid, S. Bukhari, “CNN–LSTM hybrid architecture for agricultural disease detection under temporal variability,” Comput. Electron. Agric., vol. 209, no. 107939, 2023.

F. C. A. Rahman, K. Uddin, “Comparative evaluation of machine learning techniques for olive leaf disease prediction,” Comput. Intell. Neurosci., vol. 2024, no. 991245, 2024.

F. G. J. Torres, R. García, “AI-based early detection of olive diseases for precision agriculture,” Comput. Electron. Agric., vol. 210, pp. 107–115, 2024.

N. W. A. Amin, M. Khalid, “Explainable AI framework for plant disease diagnosis using CNN and Grad-CAM,” IEEE Access, vol. 11, pp. 45902–45915, 2023.

P. C. K. Patel, R. Singh, “Efficient Net based olive disease classification with transfer learning,” IEEE Access, vol. 12, pp. 90201–90214, 2024.

M. J. A. Dali, R. Mourad, “AI-based imaging for nutrient and disease diagnosis in olive leaves,” Comput. Electron. Agric., vol. 209, pp. 107933–107948, 2023.

I. B. M. Ahmad, F. Ullah, A. R. A. Hamza, M. Usman, M. Imran, “Cotton Leaf Disease Detection Using Vision Transformers: A Deep Learning Approach,” Crops, vol. 1, no. 3, 2024.

A. Sundaraj, D. P. Isravel, and J. P. M. Dhas, “Diagnosis of Plant Leaf Disease using Vision Transformer,” Proc. 2024 10th Int. Conf. Commun. Signal Process. ICCSP 2024, pp. 82–87, 2024, doi: 10.1109/ICCSP60870.2024.10543969.

H. A. S. Najeeb, U. Rafiq, “Comprehensive analysis of CNN architectures for olive leaf disease detection,” Appl. Sci., vol. 15, no. 2, pp. 542–557, 2025.

S. A. M. Iqbal, H. Rehman, “Transfer learning-based olive leaf disease classification using EfficientNet,” IEEE Access, vol. 12, pp. 78140–78153, 20241.

K. L. Y. Li, J. Zhang, “Frequency-domain CNN for robust plant disease classification,” Comput. Electron. Agric., vol. 224, p. 109298, 2024.

Z. N. A. Javed, U. Khan, “Lightweight CNNs for mobile-based olive leaf disease detection,” Neural Process. Lett., vol. 57, pp. 1189–1202, 2024.

E. A. T. Eleyan, M. Ozturk, “Machine learning approaches for olive disease classification: Comparative analysis,” Procedia Comput. Sci., vol. 180, pp. 124–132, 2021.