• Raghavan Krishnasamy Lakshmana Perumal

Early detection of plant infections during the first weeks of growth can materially reduce yield loss and input waste. Traditional methods of plant disease detection, relying predominantly on visual inspection by trained agronomists, often identify infections only after symptoms become visibly apparent—typically 7-14 days post-infection for most viral and fungal pathogens. This delayed detection window represents a critical vulnerability in modern agricultural systems. Recent advances in multimodal artificial intelligence, combining diverse sensor modalities with machine learning architectures, offer unprecedented opportunities for pre-symptomatic disease detection systems. However, the agricultural AI community lacks standardized frameworks for developing, evaluating, and deploying multimodal detection systems. This paper proposes a builder-focused standard that covers scope and use cases, data acquisition and metadata, model architectures and fusion strategies, evaluation metrics with emphasis on early warning value, field protocols for drones and ground units and on-device inference, interoperability and data governance, risk management, and reporting. The guidance draws on accepted surveillance practice under the International Plant Protection Convention, FAIR and MIAPPE principles for plant phenotyping data, and widely used public datasets such as PlantVillage and PlantDoc.

Multimodal learning, Plant disease detection, Hyperspectral imaging, Early warning systems, Precision agriculture

"MULTIMODAL AI FOR EARLY DETECTION OF PLANT VIRAL AND PEST INFECTIONS IN EARLY GROWTH STAGES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 11, page no.a243-a261, November-2025, Available :http://www.jetir.org/papers/JETIR2511031.pdf

"MULTIMODAL AI FOR EARLY DETECTION OF PLANT VIRAL AND PEST INFECTIONS IN EARLY GROWTH STAGES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 11, page no. ppa243-a261, November-2025, Available at : http://www.jetir.org/papers/JETIR2511031.pdf