This study describes the potential use of colorimetric parameters (L*, a*, b*, C* and h*) obtained from image processing technique (IPT) to predict the contents of spoilage indicators (histamine, total volatile nitrogen and trimethylamine nitrogen) in rainbow trout fillets (600±50 g) stored at 4±1 °C during time intervals of 0, 5, 10, 15 and 20 days. Compared to 10 days, storage for 15 or 20 days provoked major changes in the colour of fish fillets, especially the C* and h* parameters (P<0.05). L* increased significantly (P<0.05) from 82.7 (day 0) to 88.18 (day 15) and then fell to 84.37 by the end of the refrigeration period. The h* decreased (P≯0.05) from 0.03 to -0.13 during the first ten days, followed by a significant (P<0.05) decline from -1.08 to -1.15 during days 15 to 20. Significant (P<0.05) multivariate regression models between spoilage indicators and colorimetric parameters (L*, C* and h*) were observed. The present study showed that IPT can be used as an auxiliary method to estimate the fillet quality.
