AgriViT-NLP: A Multi-Modal Framework for Plant Disease Detection and Farmer Query Understanding

Abstract

Abstract- Plant diseases pose a serious threat to global food security by reducing crop yield and quality. To improve detection, a multi-modal diagnostic framework is proposed that combines Vision Transformers (ViTs) for image-based disease classification with Natural Language Processing (NLP) for symptom description and treatment recommendations. The system supports multilingual interaction and generates automatic disease reports, making it accessible to diverse farming communities. By integrating ViT and NLP, the model offers higher diagnostic accuracy and interpretable, farmer-friendly support. Designed for real-world use, it can be deployed on mobile and IoT platforms, enabling smart, interactive decision-making in precision agriculture. Keywords: Plant Disease Detection, Vision Transformers, NLP, Multi-Modal Learning, Precision Agriculture.

Plant phenotyping relevance

植物画像から病害を分類するマルチモーダル診断手法が研究の中心であり、植物の病害状態を直接推定するため、フェノタイピング手法として採用する。

Code and data availability

公開本文の所在を確認できませんでした。非公開または購読が必要な可能性があります。

No evidence-backed public reproduction asset is currently recorded.