Computer Vision System: A Better Tool for Assessing Pork and Beef Colour than a Standard Colourimeter

Abstract

The aim of this paper was to evaluate the use of computer vision system (CVS) to calculate CIE colour coordinates of beef and pork, as compared to a traditional Minolta colourimeter. Statistical analysis revealed significant differences of the colour parameters (L*a*b*, hue angle and chroma) using these two different techniques for colour detection. The CVS methodology produced colours highly similar to the visual assessment tests, but the Minolta colourimeter did not. The CVS-obtained colours were similar to the colours of both pork and beef samples visualized by trained panellists on the monitor, but colourimeter-obtained colours differed. The frequency of similarity for CVS-obtained colours and the actual meat colours as seen by the trained panellists was 100%. These results indicate that the CVS could be a superior alternative over the conventional Minolta colourimeter by offering improved representativeness and accuracy. In addition to providing objective colour measurement, it offers other possibilities that can be of benefit in further quality control or research within the meat industry.