Quantification of surface iridescence in meat products by digital image analysis

Iridescence extent is commonly evaluated by sensory analysis but it is a time-consuming and cost-intensive method. A low-cost, rapid and objective alternative is digital image analysis. Here we report the development of an image analysis method for quantification of iridescence in meat products. Two segmentation techniques (global thresholding and k-means clustering algorithm) were tested for their capability to divide images into segments of iridescent and non-iridescent areas. Images segmented using k-means clustering algorithm resulted in slightly higher iridescent areas than images segmented with global thresholding (mean difference of 1.24%) but no significant difference (P > .05) between the iridescent areas calculated by both methods was observed. Almost perfect agreement (κ = 0.800, p = .001) was observed between the image analysis and the visual evaluation. The results from this study showed that digital image analysis is an effective tool for evaluating surface iridescence in meat and meat products.