Ensemble Learning Framework for Mango Plant Disease Detection and Classification

Abstract

Agriculture sector play a vital role in economy of India where the crop of mangoes is also considered as major fruit crop as it contributes significantly to the country's agricultural economy. Mango cultivation provides livelihood to millions of farmers across the country. One of the main barriers to increased food production is the diseases of the plants. Mango trees are prone to a variety of diseases and addressing them effectively can be quite challenging. This paper presents an ensemble-based classification of mango tree leaf diseases. Ensemble based classification makes use of multiple classifiers in order to make an efficient decision about the crop disease. In this paper, homogeneous VGG-19 CNN architecture is employed in bagging manner which proves the validity of the system by providing the accuracy of 95%, precision of 97%, recall of 97% and F-score of 97%.This system will be useful for Ministry of agricultural and farmer welfare for taking preventive measures to make Mango trees disease free.

Plant phenotyping relevance

マンゴー葉の画像に基づく病害分類を、アンサンブルCNNで開発・検証しており、植物の病害状態を推定する方法が中心である。

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.