The AgriGuard tool, developed under the COSA framework, was presented by Chiuzbăian Rareș as a solution for the instant identification of crop diseases.
The challenge — protecting the food supply
Plant diseases destroy up to 40% of crops worldwide, hurting farmers and raising food prices for everyone. At the same time, expert agronomists are hard to find — getting a specialist to visit a farm is slow, costly and simply not available to most farmers. The result is a critical gap between the moment a disease appears and the moment it is correctly diagnosed and treated.
The proposal
AgriGuard puts diagnostic capability directly in the hands of the farmer through a mobile-friendly tool that identifies plant diseases instantly from a single photograph. It leverages the same family of computer-vision and hybrid machine-learning models that the COSA team has been publishing on — including the Enhanced Hybrid Machine Learning Model for Plant Disease Detection and Classification (Măcelaru, Pop, Chiuzbăian, Kovacs — HAIS 2025, LNCS Vol. 16202).
Why it matters
Bringing accurate disease identification to the smartphone of every farmer enables earlier intervention, targeted treatments and reduced reliance on broad-spectrum chemicals — all of which directly improve crop yields, food safety and farm economics. AgriGuard represents the natural exploitation path for the plant-disease research line carried out within COSA Use Case 2.
Presentation slides
View the full slide deck below — or open the PDF in a new tab / download.
If the PDF does not render in your browser, please use the "open" or "download" link above.
0 Comments