AI Detection of Edible Oil Adulteration Using Sensory Attributes: Review

The adulteration of edible oils is a major global issue that impacts food safety and public health, particularly in premium oils such as groundnut, sesame, coconut, sunflower, and olive oil. Traditional detection techniques, such as chromatography and spectroscopy, offer high precision but are often lengthy and destructive. Recent developments in Artificial Intelligence (AI) have facilitated quick, non-destructive, and scalable methods for detecting adulteration. This review explores the shift from conventional techniques to AI-based approaches, which include machine learning, computer vision, and the analysis of spectroscopic data (like Near-Infrared and hyperspectral imaging). It underscores the importance of AI in evaluating sensory and physicochemical properties, such as aroma, density, and color, to enhance accuracy in detection. Additionally, the review assesses the effectiveness of various AI models, outlines the health risks associated with adulterants, and points out key challenges related to data quality, model generalization, and real-world application. Ultimately, it suggests future research avenues to develop robust, integrated, and deployable systems that ensure the authenticity and safety of edible oils.