Hyperspectral imaging-based early stress and disease detection for sustainable crop management

This review highlights how hyperspectral imaging, artificial intelligence, spectral analysis, and plant physiology enable early, accurate, non-destructive stress detection, advancing precision agriculture and sustainable food production. Early detection of plant diseases and stress is vital to sustaining crop productivity under increasingly climatic and environmental pressures. Hyperspectral imaging (HSI) enables the identification of physiological and biochemical changes in plants before visible symptoms emerge, supporting timely diagnosis and precision agriculture. This review examines the spectral mechanisms underlying plant responses to biotic and abiotic stresses across the visible, near-infrared, and shortwave infrared regions, focusing on how pigment composition, cellular structure, and water status affect spectral signatures. Here, we explain the  key aspects of hyperspectral data acquisition, calibration, and preprocessing, including radiometric correction, illumination normalization, spectral noise reduction, and dimensionality reduction, essential for analysing high-dimensional datasets. Furthermore, the integration of HSI data with machine learning and deep learning algorithms for disease classification, severity quantification, and early stress detection is also reviewed.  Therefore, this review summarizes  the potential of hyperspectral imaging and artificial intelligence for advancing precision crop monitoring and data-driven agricultural management.