Rapid determination offish freshness using colorimetric sensor array

摘要

Freshness is the most important factor in fish quality evaluation. This research was intended to develop an artificial olfaction system based on colorimetric sensor array for rapid evaluation offish freshness. Nine chemically responsive dyes includingmetalloporphyrins and pH indicator were selected according to their sensitivity to volatile compounds typical of spoilage processes in fish. The colorimetric sensor array was made by printing these selected dyes on a reverse phase silica gel plate. Fishof chub ('Hypophthalmichthys molitrix'), purchased from a local market, were detected every 24 hours beginning from two hours after death of the chubs and lasting for 7 days using a system based on the sensor array. A color change profile for each sample was obtained by differentiating the images of the sensor array before and after exposure to the analyte. The digital data sets converted from the color change images were analyzed using principal component analysis. Principal component analysis showedthe possibility of grouping chub samples into three freshness groups, corresponding to day 1, day 2-5, and day 6-7. A radial basis function neural network was employed to classify the degree of freshness, achieving an overall classification accuracy of 87.5%. This research suggests that the system can be useful for quality evaluation of fish and perhaps other food containing high protein.