Advances in Precision Systems Based on Machine Vision for Meat Quality Detection

Traditional assessment (e.g., visual inspection and biochemical analysis) is the prevailing method for meat quality assessment in the food industry. However, this approach is time-consuming, laborious, costly, and subjective. In response to the inherent limitations associated with conventional assessment, RGB (red-green-blue) cameras, hyperspectral imaging, and structured illumination reflectance imaging are gaining ample attention in the food industry. These techniques are increasingly applied to various aspects of meat quality and safety assessments, encompassing parameters such as tenderness, chemical composition, adulteration, and overall quality traits. This review focuses on scientific articles published in the past five years that leverage these machine vision techniques to address challenges in the meat processing industry. These machine-vision techniques are briefly introduced, shedding light on their principles and applications. Moreover, this review identifies the challenges and strengths associated with these technologies. To provide comprehensive insights, this review includes thoughtful solutions to overcome the challenges posed by these advanced techniques in the context of meat quality assessment within the food industry. Furthermore, we suggest a novel approach for meat processing which is integrating hyperspectral imaging with structured illumination reflectance imaging for easy detection of both surface and internal quality assessment in meat.