A Deep Learning Approach to Early Plant Disease Detection and Diagnosis Using Advanced Imaging

Abstract

Early and accurate plant disease detection is crucial for sustainable agriculture. This research proposes an Ensemble Residual Sequence Net (ERSN) architecture, combining LSTM and ResNet-50, for classifying plant diseases in pearl millet. Targeting diseases like downy mildew, ergot, blast, rust, and smut, the model achieved high accuracy (94.7%), precision, recall, F I-score, specificity, and sensitivity. This innovative approach demonstrates the potential of deep learning in enhancing plant disease management and safeguarding crop yields.

Plant phenotyping relevance

画像に基づき植物病害を分類・検出する深層学習手法の開発と性能評価が主題であり、植物の病害状態を推定するフェノタイピング手法に該当する。

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.