Adaptation regimes of pretrained neural networks in visually proximate class classification: evidence from the PlantVillage benchmark

Classifying visually proximate categories remains a difficult problem in image analysis because inter-class differences are often small, whereas intra-class variability is substantial. The aim of the study was to evaluate how the adaptation regime of a pretrained neural network affects performance in a high-complexity recognition task under a fixed architecture and identical experimental conditions. The working hypothesis assumed that partial fine-tuning of the upper layers of a pretrained convolutional neural network would outperform a regime in which the convolutional backbone remains frozen and only the final classification block is trained. The open PlantVillage dataset, containing 54,303 images and 38 classes, was used as a standardized benchmark; its subject domain was treated as a convenient testbed for complex classification of visually similar states. MobileNetV3Small served as the base model. Two adaptation regimes were compared: a frozen convolutional backbone and partial fine-tuning of the upper part of the feature extractor. The main gain was achieved with partial fine-tuning: validation accuracy increased from 0.9707 to 0.9816, while validation loss decreased from 0.0929 to 0.0576. Class-wise analysis on the independent test split showed that the F1-score exceeded 0.95 for 33 of 38 classes, whereas the lowest values, 0.8889 and 0.9078, were observed in groups with high visual similarity. The scientific novelty does not lie in the general idea of partial fine-tuning of a pretrained convolutional network, since this approach is widely used in computer vision, but in the controlled experimental comparison of two MobileNetV3Small adaptation regimes under identical conditions on the PlantVillage dataset. The results show that, for the selected architecture, fixed data split, and visually proximate class classification task, partial fine-tuning of the upper part of the feature extractor improves recognition quality and produces an interpretable error structure.

1. Aboelenin, S., Elbasheer, F. A., Eltoukhy, M. M., El-Hady, W. M., Hosny, K. M. (2025). A hybrid Framework for plant leaf disease detection and classification using convolutional neural networks and vision transformer. Complex & Intelligent Systems, 11, Article 142. https://doi.org/10.1007/s40747-024-01764-x
2. Chollet, F. (2023). Transfer learning & fine-tuning. TensorFlow Core. URL: https://www.tensorflow.org/guide/keras/transfer_learning (дата обращения: 28.03.2026).
3. Dong, J., Fuentes, A., Zhou, H., Jeong, Y., Yoon, S. et al. (2024). The impact of fine-tuning paradigms on unknown plant diseases recognition. Scientific Reports, 14, Article 17900. https://doi.org/10.1038/s41598-024-66958-2
4. Iftikhar, M., Kandhro, I. A., Kausar, N. et al. (2024). Plant disease management: a fine-tuned enhanced CNN approach with mobile app integration for early detection and classification. Artificial Intelligence Review, 57, Article 167. https://doi.org/10.1007/s10462-024-10809-z
5. Natarajan, S., Chakrabarti, P., Margala, M. (2024). Robust diagnosis and meta visualizations of plant diseases through deep neural architecture with explainable AI. Scientific Reports, 14, Article 13695. URL: https://doi.org/10.1038/s41598-024-64601-8
6. Pacal I., Kunduracioglu I., Alma M. H. [et al.]. A systematic review of deep learning techniques for plant diseases // Artificial Intelligence Review. 2024. Vol. 57. Art. 304. DOI: 10.1007/s10462-024-10944-7.
7. Rahman, M. M., Islam, M. S., Islam, M. N. (2024). Classification of Various Plant Leaf Disease Using Pretrained Convolutional Neural Network On Imagenet. The Open Agriculture Journal, 18. https://doi.org/10.2174/0118743315305194240408034912
8. Ramanjot, Mittal U., Wadhawan A. [et al.]. Plant Disease Detection and Classification: A Systematic Literature Review // Sensors. 2023. Vol. 23, No. 10. Art. 4769. DOI: 10.3390/s23104769.
9. Richter D. J., Kim K. Assessing the performance of domain-specific models for plant leaf disease classification: a comprehensive benchmark of transfer-learning on open datasets // Scientific Reports. 2025. Vol. 15. Art. 18973. DOI: 10.1038/s41598-025-03235-w.
10. Salka, T. D., Hanafi, M. B., Rahman, S. M. S. A., Zulperi, D. B. M., Omar, Z. (2025). Plant leaf disease detection and classification using convolution neural networks model: a review. Artificial Intelligence Review, 58, Article 322. https://doi.org/10.1007/s10462-025-11234-6
11. Salman, Z., Muhammad, A., Han, D. (2025). Plant disease classification in the wild using vision transformers and mixture of experts. Frontiers in Plant Science, 16, Article 1522985. https://doi.org/10.3389/fpls.2025.1522985
12. Sambana, B., Nnadi, H. S., Wajid, M. A., Fidelia, N. O., Camacho-Zuñiga, C., Ajuzie, H. D., Onyema, E. M. (2025). An efficient plant disease detection using transfer learning approach. Scientific Reports, 15, Article 19082. https://doi.org/10.1038/s41598-025-02271-w
13. Shafay, M., Hassan, T., Owais, M., Hussain, I., Khawaja, S. G. et al. (2025). Recent advances in plant disease detection: challenges and opportunities. Plant Methods, 21, Article 140. https://doi.org/10.1186/s13007-025-01450-0
14. Shafik W., Tufail A., De Silva Liyanage C. [et al.]. Using transfer learning-based plant disease classification and detection for sustainable agriculture // BMC Plant Biology. Vol. 24. Art. 136. DOI: 10.1186/s12870-024-04825-y.
15. Shoaib, M., Sadeghi-Niaraki, A., Ali, F., Hussain, I., Khalid, S. (2025). Leveraging deep learning for plant disease and pest detection: a comprehensive review and future directions. Frontiers in Plant Science, 16, Article 1538163. https://doi.org/10.3389/fpls.2025.1538163
16. (2024). plant_village. TensorFlow Datasets. URL: https://www.tensorflow.org/datasets/catalog/plant_village (дата обращения: 28.03.2026).
17. TensorFlow. (2024). tf.keras.applications.MobileNetV3Small. TensorFlow v16.1 API Documentation. URL: https://www.tensorflow.org/api_docs/python/tf/keras/applications/MobileNetV3Small (дата обращения: 28.03.2026).
18. Upadhyay, A., Chandel, N. S., Singh, K. P., Chakraborty, S. K. Deep learning and computer vision in plant disease detection: a comprehensive review of techniques, models, and trends in precision agriculture. Artificial Intelligence Review. 2025. Vol. 58. Art. 92. URL: https://doi.org/10.1007/s10462-024-11100-x
19. Yang, B., Li, M., Li, F., Wang, Y., Liang, Q., Zhao, R., Li, C., Wang, J. (2024). A novel plant type, leaf disease and severity identification framework using CNN and transformer with multi-label method. Scientific Reports, 14, Article 11664. https://doi.org/10.1038/s41598-024-62452-x
20. Zhang, H.-W., Wang, R.-F., Wang, Z., Su, W.-H. DLCPD-25: A Large-Scale and Diverse Dataset for Crop Disease and Pest Recognition. Sensors. 2025. Vol. 25, No. 22. Art. 7098. URL: https://doi.org/10.3390/s25227098
21. Zhao J., Xu L., Ma Z. [et al.]. A review of plant leaf disease identification by deep learning algorithms // Frontiers in Plant Science. 2025. Vol. 16. Art. 1637241. DOI: 10.3389/fpls.2025.1637241.