Monitoring fruit quality is vital in modern agriculture and postharvest systems due to its impact on food security, economic sustainability, and consumer health. Traditional destructive sampling and manual inspection are time-consuming and inefficient. Integrating artificial intelligence (AI) with non-destructive sensing technologies offers a rapid, high-throughput alternative for quality assessment. This review explores the application of AI across pre- and postharvest stages using spectroscopic (e.g., NIR, THz), imaging (e.g., hyperspectral, X-ray), and sensor-based systems (e.g., electronic nose). It evaluates traditional machine learning (e.g., SVM, RF), deep learning (e.g., CNNs, GANs), and hybrid models, with a focus on real-time deployment, edge AI, and multimodal data fusion. AI-driven methods have achieved over 95% accuracy in tasks like ripeness classification and bruise detection, reducing postharvest losses by 20–40%. However, challenges remain, such as high sensor costs, model fragility under field conditions, and a lack of standardization. Future directions highlight explainable AI, IoT-enabled digital twins, and autonomous agri-robotics. Overall, this review underscores AI’s transformative potential in fruit quality monitoring, contingent on interdisciplinary collaboration and scalable implementation.
Fu et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: