Remote sensing inversion of grain quality based on multi-source information: A review

Grain is an important energy source for human beings, and its quality detection and prediction are of great significance for production, storage, and sale. So far, through the comprehensive utilization of proximal, airborne, and satellite remote sensing (RS) platforms, the RS-based non-destructive inversion method of grain quality has emerged as a promising alternative to conventional chemical testing methods. Based on the in-depth analysis of literature on the multi-source RS inversion techniques for grain quality detection and prediction, this study first introduces the key quality indicators of grain seeds and the widely-used multi-scale RS platforms. Subsequently, it summarizes the methodologies employed for grain quality inversion, conducts a comparative evaluation of these methods, and ultimately identifies the deficiencies in current research endeavors as well as the potential avenues for future grain quality inversion studies leveraging multi-source information. This study provides theoretical references for latecomers to select appropriate RS techniques in grain quality detection and may offer inspiration for non-destructive estimation of physiological and biochemical parameters of other crops and varieties. • The review systematically summarise the research progress of grain quality inversion(detection & prediction). • The review compares proximal, UAV, satellite platforms for grain quality inversion. • Inversion methods（spectral features, mechanistic models, multi-source fusion）are expounded. • Challenges for the grain quality inversion are discussed.