Role of image processing and machine learning techniques in detection of crop stress and crop diseases

Abstract

India’s agriculture sector relies on satellite images-based crop stress indicators to identify crop stress and diseases, which are crucial for preventing losses. These indicators offer high spatial resolutions, low costs, and short turnaround times. Image processing and machine learning models are used to classify crops based on color, damage, area, and texture parameters. With an emphasis on potential future research approaches, this study examines popular techniques for agricultural water stress monitoring utilising image processing and machine learning. It investigates the relationship between crop drought and relative water content, equivalent water thickness, evapotranspiration, agricultural water stress, and sun-induced chlorophyll content.