Approaches to detecting peanut diseases based on remote sensing and artificial intelligence : machine learning and deep learning

Abstract

Peanuts (Arachis hypogaea L.) are a major oil seed crop grown in tropical and subtropical regions. However, each harvest is compromised by leaf diseases that affect not only yield but also quality. Traditional diagnostic methods based on visual observation have several limitations: cost, time, and technical expertise. To address these limitations, approaches based on remote sensing (multispectral and hyperspectral imaging) and artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), have been developed. Unlike previous reviews, this study presents a comparative and critical analysis of ML and DL techniques, as well as a conceptual framework integrating remote sensing and AI for the detection of peanut diseases. This analysis highlights the main limitations of current methodologies and identifies key challenges while providing directions for future research.