Classification and Prediction of Crop Diseases: A Review

Abstract

Agriculture is the foundation of civilization which faces several problems in the 21st century. Crop diseases threaten global food security by reducing yields. Visual inspection to detect diseases can be time-consuming, subjective, and error-prone. Recent advances in various machine learning (ML) and deep learning (DL) techniques have led to ease in the identification of crop diseases. ML and DL demonstrate their versatility in image recognition, segmentation, and anomaly detection. In this paper, some of the recent works based on crop-disease detection using various ML and DL techniques are reviewed. It includes early illness detection and appropriate interventions to reduce yield loss. The review emphasizes the relevance of crop disease detection for food security and sustainable agriculture. ML and DL techniques can help farmers monitor crop health and optimize resource allocation.

Plant phenotyping relevance

作物病害を植物画像から検出する機械学習・深層学習手法を主題とするレビューであり、植物の病害状態を推定するフェノタイピング手法のレビューに該当する。

Code and data availability

This is a review article summarizing prior ML/DL crop-disease studies; it presents no paper-specific phenotype datasets, images, code, or models with any availability statement or public URL. All datasets mentioned (e.g., PlantVillage, BPLD) belong to cited third-party works, not this paper.

No evidence-backed public reproduction asset is currently recorded.