A Hybrid CNN-Transformer Approach for Automated Olive Leaf Disease Classification
13th International Conference on Electrical and Electronics Engineering, ICEEE 2026, Antalya, Turkey, 27 - 29 April 2026, pp.567-571, (Full Text)
- Publication Type: Conference Paper / Full Text
- Doi Number: 10.1109/iceee69936.2026.11598274
- City: Antalya
- Country: Turkey
- Page Numbers: pp.567-571
- Keywords: convolutional neural networks (CNNs), deep learning, image classification, Plant leaf disease detection, Swin Transformer, ViT
- Yozgat Bozok University Affiliated: Yes
Abstract
In this study, a hybrid deep learning approach is proposed for the automatic classification of olive leaf diseases by integrating the complementary strengths of CNN-based architectures and the Swin Transformer. The proposed model combines the local and texture-oriented feature extraction capabilities of convolutional neural networks with the hierarchical, window-based attention mechanisms of the Swin Transformer, enabling a joint representation of both local and global visual information. The performance of the model is evaluated on an olive leaf image dataset that reflects real-world field conditions and exhibits class imbalance. Experimental results demonstrate that the proposed hybrid approach provides an effective and robust solution for the automatic diagnosis of olive leaf diseases.