Plant Disease Detection

Abstract

Agriculture remains a critical pillar of economic sustainability, particularly in developing countries where crop productivity directly impacts food security and farmer income. However, plant diseases significantly reduce crop yield and quality, leading to substantial economic losses. Traditional disease detection techniques rely on manual inspection and expert consultation, which are time-consuming, subjective, and often inaccessible in rural areas. This paper presents an advanced deep learning-based system for plant disease detection, severity classification, and outbreak prediction. The proposed framework integrates Convolutional Neural Networks (CNNs) for image-based disease identification, clustering techniques for severity classification, and Long Short-Term Memory (LSTM) models for time-series-based prediction. The system also incorporates feature importance analysis using Random Forest to enhance interpretability. A web-based interface enables real-time interaction, allowing users to upload images and receive immediate diagnostic feedback. Experimental results demonstrate high accuracy and robustness across diverse datasets. The integration of multiple learning paradigms ensures improved performance, scalability, and adaptability, making the system suitable for real-world agricultural applications

Code and data availability

The paper describes a CNN/K-means/LSTM plant disease detection system but provides no public dataset, code, model, or supplement of its own. It only mentions using publicly available sources such as PlantVillage (cited prior work, not a paper-specific asset), and contains no availability statement or URL for theauthors

No evidence-backed public reproduction asset is currently recorded.