Multi-Organ Plant Disease Detection Using CNN and Machine Learning: A Root-to-Leaf Approaches

Abstract

Detecting crop diseases is critical but labor-intensive task in agriculture, often requiring expert knowledge and manual inspection. This paper describes an efficient technique for automated disease using computer vision and Machine learning. The system analyzes images of plant leaves, stems, and roots to identify symptoms with high accuracy using Otsu's thresh-olding. A structured data acquisition process ensures quality input, while convolutional neural networks (CNNs) enable robust classification. This approach reduces reliance on skilled labor, supports early disease intervention, and improves overall crop health monitoring. The solution is designed for scalability and real-time use, including mobile-based applications for on-field diagnosis.

Plant phenotyping relevance

植物の葉・茎・根の画像から病徴を自動検出する画像解析・CNN手法が研究の中心であり、植物病害状態の表現型推定に該当する。

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.