“AgriFutura”-Plant Disease Detection with Fertilizer Recommendation using Deep Learning (CNN)

Abstract

Abstract - Agriculture is an important, critical sector that ensures the provision of sustenance and economic viability. Among the critical issues that affect agricultural practitioners is the ability to detect plant diseases in real time, which has critical impacts on plant yield and viability. Traditionally, disease detection requires human assessment and consultation, which is associated with costs and human error. To overcome the aforementioned limitations and shortcomings, this work introduces AgriFutura, an online plant disease identification and fertilizer guidance platform that utilizes deep learning concepts. The proposed platform utilizes a Convolutional Neural Network (CNN) architecture to evaluate images of plant leaves and identify them based on health and disease. In addition to healthcare guidance and recommendations, the platform also provides disease-oriented fertilizer guidance and recommendations, which can aid decision-making. The proposed platform is simple to use and therefore does not incur costs that are normally associated with specialized technical skills and knowledge. Results from experimental studies indicate that the proposed platform is effective and has short response times, which therefore qualifies to be used in real-life agricultural settings. Key Words: Smart Agriculture, Plant Disease Detection, Convolutional Neural Network, Deep Learning, Fertilizer Recommendation

Plant phenotyping relevance

葉画像から植物の健康状態・病害をCNNで推定するプラットフォームが研究の中心であり、植物病害状態の画像ベース表現型計測に該当する。肥料推薦も含むが、病害検出手法自体が主要な貢献である。

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.