Comparison of different image analysis algorithms on MRI to predict physico-chemical and sensory attributes of loin

Computer vision algorithms on Magnetic Resonance Imaging have been introduced as an alternative to destructive methods for determining the quality traits of meat products, since Magnetic Resonance Imaging is a non-destructive, non-ionizing, and innocuous method. The use of fractal properties to analyze Magnetic Resonance Imaging could be another possibility for this purpose. In this paper, a new fractal properties algorithm was developed, to obtain features from Magnetic Resonance Imaging based on fractal characteristics. This algorithm is called One Point Fractal Texture Algorithm. The fractal properties of this new algorithm were compared with those of other fractal (Classical Fractal Algorithm and Fractal Texture Algorithm) and the classical texture ones (Grey level co-occurrence matrix, grey level run length matrix and neighbouring gray level dependence matrix). The results obtained by means of these computer vision algorithms were correlated to the results obtained by means of physico-chemical and sensory analysis. Classical Fractal Algorithm reached low association for the quality parameters of loins. The remaining algorithms achieved correlation coefficients higher than 0.5, noting One Point Fractal Texture Algorithm that reached the highest correlation coefficients in all cases except for the L* coordinate color. These high correlation coefficients confirm the new algorithm as an alternative to the other computer vision approaches in order to compute the physico-chemical and sensory parameters of meat products in a non-destructive and efficient way.