Integration of Hyperspectral Imaging with Machine Learning for Quality Assessment of Nuts: A Systematic Review

Nuts such as pecans, almonds, peanuts, pistachios, and walnuts are nutrient-dense foods rich in unsaturated fatty acids and antioxidant compounds. Their regular consumption has been linked to significant health benefits, including reduced risks of cardiovascular disease, diabetes, and high cholesterol. With increasing global demand, ensuring the quality of nuts before they reach consumers is critical. Conventional quality assessment methods dominate the industry but are often subjective, destructive, time-intensive, environmentally burdensome, and laborious. Therefore, there is an urgent need for rapid, non-destructive, and objective alternatives capable of meeting modern quality standards. In this systematic review, we summarize traditional approaches for evaluating nut quality parameters and introduce hyperspectral imaging as a novel technique with promising applications. We examine its use in detecting nut adulteration, assessing chemical composition, identifying defects, and evaluating other quality traits. Limitations of hyperspectral imaging in industrial settings are also discussed, along with potential solutions and future directions. Given the relatively limited research area, approximately 44 relevant studies were critically reviewed. This work provides valuable insights for researchers and industry stakeholders developing innovative technologies for nut quality assessment.