Peer-Reviewed Journal Details
Mandatory Fields
Sheridan, C; O'Farrell, M; Lewis, E; Flanagan, C; Kerry, J; Jackman, N
Journal Of Optics A-Pure And Applied Optics
A comparison of CIE L*a*b* and spectral methods for the analysis of fading in sliced cured ham
Optional Fields
optical fibres spectral reflectances principal component analysis multi-layer perceptron CIE L*a*b* sliced cured ham ham surface colour fading NEAR-INFRARED SPECTROSCOPY COLOR STABILITY CHILL STORAGE OPTIMIZATION MEAT
In the modern retail environment, the appearance of a product is frequently the only quality indicator available to consumers. This is especially true of products such as sliced ham that have been sealed into packages to maintain product freshness. It has been shown that sliced ham products undergo discolouration from their original pink colour to a pale grey colour when exposed to a combination of oxygen and light. This is unappealing to consumers who expect a pink colour for sliced ham. An investigation is made into a sensor that would monitor the initial colour status of cured ham before packaging in order to determine the amount of time left before the ham fades to an unsatisfactory colour. For this sensor to operate, appropriate analysis of the appearance of the meat is required. Two methods for the measurement of the fading were investigated - CIE L* a* b* measurements and analysis of the spectral reflectance of the colour of the ham. Several sliced ham products with differing amounts of fading were examined using both methods. It was observed that the products used covered a wide range of variation in colour. Reproducibility of CIE L* a* b* values proved to be quite difficult and significant overlapping of the L* ( lightness) and a* ( redness) values measured for pink and grey coloured ham was observed. The variations in these values can be attributed to differences in the intensity of reflected light for different products. L* a* b* measurements are sensitive to light intensity and pigment concentration. Analysis of the spectral reflectance readings did not encounter these problems as the spectral response was normalized ( to reduce intensity errors) before data analysis was carried out on the spectral shape or 'pattern' using principal component analysis ( PCA) and artificial neural networks ( ANN). A classifier based on PCA and ANN was successfully implemented that can discriminate different stages of fading for the ham slices. A case study was carried out on ham slices that had two different initial colours - light and dark. The results of the case study show that the sensor system can better discriminate between a light initial colour and a dark faded colour than the CIE L* a* b* colour measurement system.
Grant Details