AI&ML-based smart non-disruptive methods for post-harvest fruit grading: State-of-the-Art

Intelligent vision-based non-destructive evaluation solutions are built around visual criteria, which are fundamentally similar to the cognitive processes of looking, experiencing, watching, and making decisions in human perception. Among these factors, image processing stands out as a popular method for effective evaluation and grading, providing significant advantages in terms of time, expense, and resource use in a variety of fields, including agriculture. This study provides an extensive overview of state-of-the-art computer vision methods for nondestructive post-harvest fruit grading, with an emphasis on Machine Learning and Deep Learning techniques. It clarifies the changing terrain of visual assessment techniques by combining a variety of research endeavors. This study aims to present a comprehensive overview of the most recent methods, their uses, and their efficiency in fruit grading applications. This paper’s main contribution is its methodical synthesis and analysis of the body of literature, which results in a tabular overview that emphasizes the different approaches and their significance to the area. By outlining the challenges, the paper directs future research efforts and paves the way for improvised fruit grading systems that are more precise and productive.