The adulteration of spices presents a substantial risk to public health and the authenticity of culinary customs. This article introduces a comprehensive interdisciplinary strategy to tackle this problem, by combining traditional and advanced techniques to authenticate spices in simple manner.The machine learning and artificial intelligence, has shown shows significant promise in diverse area based on their ability to differentiate on the basis of distinct morphological features. This article oversees the possibility to explore machine learning to check the quality of the spices and to identify their adulterants.
As a representative case study, machine learning models for spice recognition, including transfer learning with pre-existing models such as MobileNetV2 is reported for the selected spice (Capsicum annum) demonstrating 97.97% accuracy.
In addition, the study examines several conventional and analytical methodologies viz., qualitative tests, microscopy, colorimetry, density measurement, and spectroscopy. Furthermore, quantitative method based on high-performance liquid chromatography was developed and reported here briefly for analysing the capsaicin concentration, which is essential for assessing the quality of C. annum.
This comprehensive study focuses on important areas of research that have not been well explored. The results may open a new vista in determining the quality of spices and identification of their adulterations using the artificial intelligence.