A Review of Machine Learning and Deep Learning Techniques for Saffron Adulteration Prediction System

One of the most expensive spices, saffron, commonly known as the “golden spice,” is grown primarily in Iran, Greece, Spain, China, and India (mostly in Kashmir valley). However, because of its low output and high cost, it is frequently contaminated, which causes financial loss and a number of health problems. Sunflower, safflower, maize outer layer, and hibiscus stigmas are some of the adulterants that are combined with authentic saffron samples to boost its yield. Nowadays, the menace of saffron adulteration is one of the major issues that needs further attention as human vision fails to discriminate between pure and adulterated saffron samples. Hence, an intelligent computer vision-enabled saffron adulteration prediction system is used by researchers to overcome this issue. In this paper, we present a novel framework for saffron adulteration prediction using intelligent data-driven approaches. Moreover, various pre-processing techniques are used to enhance the quality of saffron images. Besides, we demonstrate the comparative analysis of recent saffron adulteration prediction techniques and identify certain open research issues that needs futuristic attention. In addition, performance protocols such as performance metrics and saffron images-based dataset are briefly illustrated. Our analysis reveals that this study provides researchers with new directions that needs to be addressed in future and can potentially prevent economic loss and ensure consumer safety.