Automated Detection of Groundnut Plant Leaf Diseases using Convolutional Neural Networks

Abstract

Background: Groundnuts, commonly known as peanuts, are a significant legume cultivated globally, with China and India being the leading producers. Groundnut production faces challenges from pests, diseases and climate change, impacting yield and quality. In addition to notable diseases like rust, early and late leaf spots, plant health is also impacted by nutritional deficiencies. Sustainable production requires the management of diseases and the supply of appropriate nourishment. Methods: This study applies machine learning (ML), specifically Convolutional Neural Networks (CNNs), to detect groundnut disease symptoms. A CNN-based tool is developed to assess disease severity efficiently. The model is trained and validated using a dataset of 3,058 groundnut leaf images, classified into five disease categories. Result: After 100 epochs, the CNN model reached a training accuracy of 91.94% and a validation accuracy of 90.97%. Performance metrics such as precision, recall and F1-score confirm the model’s effectiveness in disease classification. The study acknowledges certain limitations, including a small dataset and a focus only on leaf infections. Future work may expand the dataset, include other plant parts and compare various ML approaches.

Plant phenotyping relevance

CNNで葉の病徴を直接解析し、病害カテゴリと重症度を推定する手法を開発・検証しており、植物表現型取得が中心です。

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.