Implementation for Plant Disease Classification via Telegram

Abstract

This study aims to develop an automated system for classifying vegetable plant diseases using the MobileNet algorithm integrated with a Telegram Bot. The system is designed to assist users, especially farmers, in identifying plant diseases quickly and efficiently through leaf images. The research method applies a Convolutional Neural Network with the MobileNet architecture due to its lightweight and efficient computational performance. The dataset used in this study consists of tomato leaf images obtained from a public dataset on Kaggle, which includes several disease categories and healthy leaves. The system is implemented using Python and integrated with the Telegram Bot API to enable real-time interaction. The process begins when users upload leaf images, followed by image preprocessing and classification using the trained model. The results show that the system is capable of providing accurate classification with good performance and can handle various input conditions. In addition, the integration with Telegram makes the system easily accessible without requiring additional applications. Therefore, this study offers a practical and efficient solution for early detection of plant diseases using deep learning technology.

Plant phenotyping relevance

葉画像から植物病害を分類する深層学習システムの開発が研究の中心であり、植物の病害状態を直接推定する画像ベースのフェノタイピング手法に該当する。

Code and data availability

The paper uses a public Kaggle tomato leaf image dataset and a MobileNet/Telegram bot implementation, but provides no specific Kaggle dataset identifier or URL, and no availability statement or public link for the authors' code, trained model, or workflow. No paper-specific, actionable public asset is identified.

No evidence-backed public reproduction asset is currently recorded.