Freshness assessment of gilthead sea bream (Sparus aurata) by machine vision based on gill and eye color changes

The fish freshness was evaluated using machine vision technique through color changes of eyes and gills of farmed and wild gilthead sea bream (Sparus aurata), being employed lightness (L^*), redness (a^*), yellowness (b^*), chroma (c^*), and total color difference (ΔE) parameters during fish ice storage. A digital color imaging system, calibrated to provide accurate CIELAB color measurements, was employed to record the visual characteristics of eyes and gills. The region of interest was automatically selected using a computer program developed in MATLAB software. L^*, b^*, and ΔE of eyes increased with storage time, while c^* decreased. The a^* parameter of fish eyes did not show clear a trend with storage time. The L^*, b^*, and ΔE of fish gills increased with storage time, but a^* and c^* decreased. Regression analysis and artificial neural networks approaches were used to correlate the eyes and gills color parameters with the time of storage and a strong correlation was found between color parameters and storage day. Gills color changes were more precise than those found for eyes in order to evaluate the fish freshness. However, the gills cover should be removed for taking the images and thus, the method is destructive and time-consuming. Therefore, the color parameters of fish eyes can be used as a green, low cost and easy method for fast and on-line assessing of fish freshness in food industry.