A novel colorimetry analysis used to compare different drying fish processes

The colour of dried fish is a measure of their biochemical and structural degradation. In this study a novel method for measuring colour by image processing is proposed which can be used in the `L∗a∗b∗ʹ space to determine the mean colour and standard deviation of heterogeneous samples with a high degree of consistency and reproducibility. Image processing of the fish samples coupled with a statistical (discriminant) analysis could separate the effect of the drying processes on the mean colour from their effect on the heterogeneity of colour. Finally, chroma variance was shown to be the only way of statistically separating (by analysis of variance) the samples according to the drying technique used. Compared with sensory analysis, measuring colour by image processing was therefore found to be reliable and efficient and could easily be integrated into an industrial production process.