The paper uses modern artificial intelligence methods to investigate models and methods for determining grape disease. The existing methodologies for classification and recognition by images of grape diseases using neural networks are analyzed. Several problems for improving recognition results are highlighted. A key component of the study was the creation of extensive datasets, training neural network models with reduced computational requirements (MobileNetV3_Large, EfficientNet_B1, and ShuffleNetV2_x2), optimizing them for the mobile environment, and integrating them into a cross- platform application using the ReactNative framework. For the study, two main datasets were used: on plant diseases(PlantVillage, PlantDoc) and grape diseases (IDADP, IPM, DownyMildew, ESCA, PlantVillage, data from open search engines). The sets have been thoroughly cleaned and balanced using artificial data augmentation techniques such as rotation, scaling, contrast change, and lighting using the PlayTorch library. Training and testing of neural networks were carried out using the PyTorch library. Transmission learning was applied to improve the performance and accuracy of the models. The models showed high accuracy scores (over 93 %) with accuracy, completeness, and F1-score scores above 85 %, and the ensemble model, combining the predictions of the three architectures, demonstrated an accuracy of about 95 %.
A mobile application has been developed for iOS and Android devices, which provides identification of grape disease based on images of leaves and bunches. The application is focused on convenience and accessibility: an intuitive interface allows you to scan in real time, view and save results, and receive treatment recommendations. Using the prototype of the developed software system using the React Native framework, an analysis of the performance of models was carried out on different types of devices, which made it possible to assess their efficiency and stability in real-world application scenarios. The advantage of the mobile application is the ability to use it in vineyards without Internet access.
In conclusion, the study’s main results are presented, and a decision is made about the possibility of using neural networks and an ensemble model to develop and operate a cross-platform mobile software system for recognizing grape diseases.
[1] Atallah, S. S., Gomez, M. I., Fuchs, M. F., & Martinson, T. E. (2011). Economic Impact of Grapevine Leafroll Disease on Vitis vinifera cv. Cabernet franc in Finger Lakes Vineyards of New York. American Journal of Enology and Viticulture, 63(1), 73–79. https://doi.org/10.5344/ajev.2011.11055
[2] Paymode, A. S., & Malode, V. B. (2022). Transfer Learning for Multi-Crop Leaf Disease Image Classification using Convolutional Neural Network VGG. Artificial Intelligence in Agriculture, 6, 23–33. https://doi.org/10.1016/j.aiia. 2021.12.002
[3] Diana Andrushia, A., Mary Neebha, T., Trephena Patricia, A.,
Umadevi, S., Anand, N., & Varshney, A. (2022). Image- based disease classification in grape leaves using
convolutional capsule network. Soft Computing. https://doi.org/10.1007/ s00500-022-07446-5
[4] Z. Huang, A. Qin, J. Lu, A. Menon and J. Gao (2020). Grape Leaf Disease Detection and Classification Using Machine Learning. International Conferences on Internet of Things
(iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) and IEEE Congress on Cybermatics (Cybermatics), 870–877.
https://doi.org/10.1109/iThings-GreenCom-CPSCom- SmartData-Cybermatics50389.2020.00150
[5] Liu, B., Tan, C., Li, S., He, J., & Wang, H. (2020). A Data Augmentation Method Based on Generative Adversarial Networks for Grape Leaf Disease Identification. IEEE Access, 8, 102188–102198. https://doi.org/10.1109/access.2020. 2998839
[6] Nader, A., H.Khafagy, M., & Hussien, S. A. (2022). Grape Leaves Diseases Classification using Ensemble Learning and Transfer Learning. International Journal of Advanced Computer Science and Applications, 13(7). https://doi.org/ 10.14569/ijacsa.2022.0130767
[7] Yin, X., Li, W., Li, Z., & Yi, L. (2022). Recognition of grape leaf diseases using MobileNetV3 and deep transfer learning. International Journal of Agricultural and Biological Engineering, 15(3), 184–194. https://doi.org/10.25165/ j.ijabe. 20221503.7062
[8] Ahmed, H. A., Hama, H. M., Jalal, S. I., & Ahmed, M. H. (2023). Deep Learning in Grapevine Leaves Varieties Classification Based on Dense Convolutional Network. Journal of Image and Graphics, 11(1), 98–103. https://doi.org/ 10.18178/ joig.11.1.98-103
[9] Huang, Y., Li, N., & Liu, Z. (2023). An Enhanced Convolutional Neural Network for Accurate Classification of Grape Leaf Diseases. Information Dynamics and Applications, 2(1), 8–18. https://doi.org/10.56578/ida020102
[10] Gupta, J., Pathak, S., & Kumar, G. (2022). Deep Learning (CNN) and Transfer Learning: A Review. Journal of Physics: Conference Series, 2273(1), 012–029. https://doi.org/ 10.1088/1742-6596/2273/1/012029
[11] Hughes, D. P., & Salathe, M. (2016). An open access repository of images on plant health to enable the development of mobile disease diagnostics. arXiv. https://doi.org/ 10.48550/ arXiv.1511.08060
[12] Singh, D., Jain, N., Jain, P., Kayal, P., Kumawat, S., & Batra, N. (2020). PlantDoc. У CoDS COMAD 2020: 7th ACM IKDD CoDSand 25th COMAD. ACM. https://doi.org/ 10.1145/ 3371158.3371196
[13] YUAN, Y., & CHEN, L. (2022). An image dataset for IDADP-grape disease identification. China Scientific Data, 7(1). https://doi.org/10.11922/11-6035.csd.2021.0077.zh
[14] Fruits and Nuts : Grape. IPM Images: The Source for Agriculture and Pest Management Pictures. https://www. ipmimages.org/browse/Areathumb.cfm?area=75
[15] Abdelghafour, F., Keresztes, B., Deshayes, A., Germain, C., & Da Costa, J.-P. (2021). An annotated image dataset of downy mildew symptoms on Merlot grape variety. Data in Brief, 37, 107250. https://doi.org/10.1016/j.dib.2021. 107250
[16] Alessandrini, M., Calero Fuentes Rivera, R., Falaschetti, L., Pau, D., Tomaselli, V., & Turchetti, C. (2021). A grapevine leaves dataset for early detection and classification of esca disease in vineyards through machine learning. Data in Brief, 35, 106809. https://doi.org/10.1016/j.dib.2021.106809
[17] Howard, A., Sandler, M., Chen, B., Wang, W., Chen, L.-C., Tan, M., Chu, G., Vasudevan, V., Zhu, Y., Pang, R., Adam, H., & Le, Q. (2019). Searching for MobileNetV3. У 2019 IEEE/CVF International Conference on Computer Vision (ICCV). IEEE. https://doi.org/10.1109/iccv.2019.00140
[18] Tan, M. and Le, Q. V. (2019) EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the 36th International Conference on Machine Learning, ICML 2019, 6105-6114. https://doi.org/10.48550/ arXiv. 1905.11946
[19] Ma, N., Zhang, X., Zheng, H.-T., & Sun, J. (2018). ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design. У Computer Vision – ECCV 2018 (с. 122–138). Springer International Publishing. https://doi.org/ 10.1007/978-3-030-01264-9_8
[20] Mohammed, A., & Kora, R. (2023). A Comprehensive Review on Ensemble Deep Learning: Opportunities and Challenges. Journal of King Saud University - Computer and Information Sciences. https://doi.org/10.1016/j.jksuci. 2023.01.014
[21] Jia Deng, Wei Dong, Socher, R., Li-Jia Li, Kai Li & Li Fei- Fei. (2009). ImageNet: A large-scale hierarchical image database, 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. https://doi.org/10.1109/ cvprw.2009.5206848