English

Image-Based Classification of Olive Varieties Native to Turkiye Using Multiple Deep Learning Architectures: Analysis of Performance, Complexity, and Generalization

Computer Vision and Pattern Recognition 2026-02-24 v1

Abstract

This study compares multiple deep learning architectures for the automated, image-based classification of five locally cultivated black table olive varieties in Turkey: Gemlik, Ayvalik, Uslu, Erkence, and Celebi. Using a dataset of 2500 images, ten architectures - MobileNetV2, EfficientNetB0, EfficientNetV2-S, ResNet50, ResNet101, DenseNet121, InceptionV3, ConvNeXt-Tiny, ViT-B16, and Swin-T - were trained using transfer learning. Model performance was evaluated using accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), Cohen's Kappa, ROC-AUC, number of parameters, FLOPs, inference time, and generalization gap. EfficientNetV2-S achieved the highest classification accuracy (95.8%), while EfficientNetB0 provided the best trade-off between accuracy and computational complexity. Overall, the results indicate that under limited data conditions, parametric efficiency plays a more critical role than model depth alone.

Keywords

Cite

@article{arxiv.2602.18530,
  title  = {Image-Based Classification of Olive Varieties Native to Turkiye Using Multiple Deep Learning Architectures: Analysis of Performance, Complexity, and Generalization},
  author = {Hatice Karatas and Irfan Atabas},
  journal= {arXiv preprint arXiv:2602.18530},
  year   = {2026}
}
R2 v1 2026-07-01T10:45:10.585Z