Measuring Color

Choice of instrument depends on the food and the application type. Color discrimination threshold of the human eye greatly differs from the color differences defined by CIE. Using CIE values, color modeling has been developed for specific applications. Reflectance data can be reported as CIE L*a*b values: L – Light, a* – red, and b* – yellow.

Color Modeling in Fruits and Vegetables

Research attempts have been made to model color values. For example, vegetables when over-blanched can change to a green color. Depending on chlorophyll and chlorophyllide destruction, a generalized model for vegetables could be found. Chromatic changes of broccoli under modified atmosphere packaging at 20 degrees Celsius in perforated and unsealed polypropylene film packages for a storage period of 10 days indicated that using L*c*h* color space diagram, the modified atmosphere generated inside the perforated film packages with 4 macro-holes was the most suitable in maintaining the chromatic quality of the broccoli heads (Rai et al. 2009).

An important parameter of the postharvest life of tomatoes is color. One color model correlates the color level and biological age at harvest (Schouten et al. 2007). Data were analyzed using non-linear regression analysis and found that biological age of tomatoes can well be predicted at farmers’ level and can save a lot of postharvest losses. Interestingly, they also found a very good correlation between the color values and tomato firmness.

Precision of prediction using models having the parameters of a, b, and their product (a×b) was verified by sensory evaluation of 55 ripe mangoes. It was found that the fruits predicted to be mature could ripe with high-satisfied taste, while the ones predicted to be immature or over mature were mostly rejected by the panels (Jha et al. 2007). Hence, these mathematical relationships between ripeness, overall quality, and freshness index can be calculated.

The relationship between color parameters and anthocyanins of four sweet cherry cultivars using L*, a*, b*, chroma, and hue angle parameters (Berta et al. 2007) indicated that chromatic functions of chroma and hue correlate closely with the evolution of color and anthocyanin levels during storage of sweet cherries.

It was also shown that color measurements can be used to monitor pigment evolution and anthocyanin content of cherries.

The above paragraphs indicate that significant attempts have been made to model color values or combination thereof for prediction of various surface, as well as internal quality parameters, of various fruits and vegetables. However, very limited work on modeling of color values of other foods, such as food grain and oilseeds, are reported for prediction of their quality parameters. The coefficient of determination of these models may not always be as high as expected. In such cases, one may try to obtain the complete spectra of specimen instead of individual color values (L*, a*, b*, etc.) in the visible range of wavelength (400 to 700 nm) and develop models using the absorption or reflectance data.

Hue value—which identifies whether an object is red, yellow, green, or blue—research is underway and new equipment is being invented to address hue values. With more research underway and companies investing in color detection instrumentation, the visible color differences observed during stress of drought, heat, or other deficiencies or development of fruits will be possible in the near future. Subtle differences in color and purchasing decisions will be taken as a marketing advantage.

Dr. Veeramuthu works at American Licorioce Company as a senior QA manager and can be reached at 219-324-1464 or [email protected].

References Furnished Upon Request