IoT Enabled Smart Agriculture System for Detection and Classification of Tomato and Brinjal Plant Leaves Disease

Abstract

Internet of Things (IoT) assisted smart farming techniques are gradually being used efficiently for identification and classification of vegetable plant diseases. Detection and classification of diseases in these plant families like Solanaceae are still problematic using DCNN due to variations in environmental conditions, genome variation, type of disease, etc. In this paper, two methods for spotting and diagnosing diseases of brinjal and tomato plants leaves named as Optimal Environmental Traversing Alert (OETA) and Optimum diagnosis of Solanaceae leaf diseases (ODSLD) respectively have been proposed. The OETA machine learning (ML) based method is used first to detect the disease, and then the ODSLD deep convolutional neural networks (DCNN) method is used to classify it. An analysis of the proposed method experiments showed that OETA disease detection for brinjal plant (eggplants) was 97.81 percent and for tomato plants was 99.03 percent. For disease classification by ODSLD method, the VGG-16 for brinjal plant and ResNet-50 for tomato plants outperformed other existing DCNN computer vision methods.

Plant phenotyping relevance

トマトおよびナス葉の病徴を画像・機械学習で検出／分類する手法の提案と性能評価が中心であり、植物の病害状態を直接推定するフェノタイピング手法に該当する。

Code and data availability

植物フェノタイピング解析を再現する公開資産であることを、入力本文と直接リンクから確認できなかったため保留しました。

No evidence-backed public reproduction asset is currently recorded.