EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Integrated Agro‑Diagnostic Framework with Deep Learning for Crop Disease Management

V. Lokeswara Reddy, V. Srilatha, V. Ganga Devi, P. Shiva Durga, R. Tharun, T. Harsha, Vardhan Reddy

Journal of Information and Communications Technology Algorithms Systems and Applications · 2026

DOI: 10.64189/ict.26320 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Information and Communications Technology Algorithms Systems and Applications, 2026. Verfasst von V. Lokeswara Reddy und 6 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Plant diseases are a constant issue in food production, resulting in worldwide losses of 20–40% on an annual basis and more than $147 billion in damages. In India, more than 150 million farmers have an annual contribution of nearly 18% to the country's GDP, and timely diagnosis of diseases is not easily accessible in rural areas because of the very limited availability of trained plant pathologists, which leads to the loss of yields in the field ranging between 25% and 50%. Most of the existing CNN-based systems are only concerned with the detection of diseases, and farmers are not provided with any practical suggestions for treatment or input procurement. In response, we present a lightweight CNN network called AgriNet, trained on the PlantVillage dataset of 61,486 labeled leaf images that represent 38 different classes for 14 different crops, and present an Integrated Agro-Diagnostic and Remediation Framework (IADRF) to address this. Under an identical experimental setup (same dataset, data split and training protocol), AgriNet is compared with Xception, ResNet-50 and Inception V3 in terms of accuracy, precision, recall, F1-score, number of parameters and inference latency. The proposed model not only reaches high accuracy of 98.68% but also yields a macro-averaged F1-score of 98.47% while using far fewer trainable parameters than the conventional CNN architectures. The prototype IADRF combines disease predictions with a structured collection of information regarding disease prevention and management and presents it through a user-friendly web interface on a mobile device. The reported latency results prove the possibility of calculations on the server side and in low resource areas, but direct deployment of the smartphone and validation at the field level remains a topic for further research.

Zitieren

APA

Reddy, V. L., Srilatha, V., Devi, V. G., Durga, P. S., Tharun, R., Harsha, T., & Reddy, V. (2026). Integrated Agro‑Diagnostic Framework with Deep Learning for Crop Disease Management. Journal of Information and Communications Technology Algorithms Systems and Applications. https://doi.org/10.64189/ict.26320

Vancouver

Reddy VL, Srilatha V, Devi VG, Durga PS, Tharun R, Harsha T, et al. Integrated Agro‑Diagnostic Framework with Deep Learning for Crop Disease Management. Journal of Information and Communications Technology Algorithms Systems and Applications. 2026. doi:10.64189/ict.26320

BibTeX
@article{Reddy2026,
  author = {Reddy, V. Lokeswara and Srilatha, V. and Devi, V. Ganga and Durga, P. Shiva and Tharun, R. and Harsha, T. and Reddy, Vardhan},
  title = {Integrated Agro‑Diagnostic Framework with Deep Learning for Crop Disease Management},
  journal = {Journal of Information and Communications Technology Algorithms Systems and Applications},
  year = {2026},
  doi = {10.64189/ict.26320}
}
Kostenlos starten →