Can the Image Processing Technique Be Potentially Used to Evaluate Quality of Frying Oil?

Abstract

The objective of this study was to investigate the feasibility of a computer vision system (CVS) for assessing the contact angle of frying oil. The oil was used to fry carbohydrate- and protein-based foods for 40 h, and the oil was collected for measuring free fatty acids (FFA), peroxide value (PV), total polar materials (TPMs), and FOS reading (dielectric constant). The results showed that FFA linearly increased with frying time (R2 > 0.95) while the polynomial correlation between TPMs and FOS reading as a result of time was observed (R2 > 0.97). The contact angle obtained from CVS was highly correlated with all chemical qualities (R2 > 0.94), except PV. In addition, the contact angle models could be used to adequately predict FFA, TPMs, and FOS reading of frying oil (R2 > 0.91). This result suggested that the image processing technique through CVS could be an appropriate alternative to chemical analysis, especially for small- and medium-scale industrial frying.