Back

Detecting color differences in lentil flour samples using an inexpensive, hand-held colorimeter.

Viscarra, S.; Vargas, A.; Jardine, L.; Tulbek, M.; Bett, K.

2025-07-10 systems biology
10.1101/2024.04.22.590631 bioRxiv
Show abstract

Our lentil breeding program was interested in determining the effects of the red lentil genotype in processed food products colour intensity. This required a colorimeter that could accurately assess colour from the CIE L*a*b* colour space on a small sample of milled flour. Flour colour is commonly assessed with industry standard HunterLab colorimeters that require a large sample size ([~]6g flour/sample). In recent years a new spectrophotometer (Nix) has become available. It was designed to quickly and accurately obtain CIEL*a*b* values for everyday items and requires a small amount of sample for analysis (0.5 g flour/sample), but its use has not been tested in milled flour products. To test whether this instrument was appropriate for milled flour colour, eight red lentil genotypes were evaluated for red colour using both instruments. Mean colour scores with Nix (a* and b*) were comparable to those obtained with the HunterLab (r2 = 0.9-0.95), L* scores were less comparable (r2 = 0.58-0.61). Both instruments were able to differentiate and rank red colour intensity in milled flour from the different red lentil genotypes. Nix procedures are more repeatable and use less sample than HunterLab validating its use for assessing red colour intensity in lentil flour. Research HighlightsO_LICompanies making foods using red lentil flour seek red lentil genotypes that exhibit red colour intensity. C_LIO_LINix Spectro2 is an inexpensive photospectrometer made for small volume samples. It has not been tested on flour products. C_LIO_LIThe selected red lentil genotypes tested showed a range of red colour intensities. C_LIO_LINix Spectro2 performance was comparable to industry standard HunterLab colorimeter. C_LI

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.