Potential of flatbed scanner for evaluation of flour samples for dark specks and flour color

Authors: Brabec, Daniel - Dan; Grothe, Sophia; Perez-Fajardo, Mayra; Pordesimo, Lester; Yeater, Kathleen

Submitted to: Cereal Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/19/2023
Publication Date: 1/29/2024
Citation: Brabec, D.L., Grothe, S.M., Perez-Fajardo, M.A., Pordesimo, L.O., Yeater, K.M. 2024. Potential of flatbed scanner for evaluation of flour samples for dark specks and flour color. Cereal Chemistry. 101:508–517. https://doi.org/10.1002/cche.10758.
DOI: https://doi.org/10.1002/cche.10758

Interpretive Summary: As wheat flour is being produced, it is important to identify ways to evaluate flour quality rapidly and reliably. For example, milling of wheat grain contaminated with smut from fungi can cause flour to have a grayish tint and carry off-odors, which collectively reduces market value. Commercial flatbed scanners combined with image analysis software have the potential to detect and quantify impurities in flour, including smut. The current study aimed to determine whether this approach could be used to detect black specks associated with various levels of smut contamination. Although the resolution of the images collected by the scanner greatly impacted numerical counts of black specks and resulted in variation, calculating percent area smut (defined as the count of speck pixels relative to the total number of image pixels) resulted in more consistent measurements of smut contamination. These measurements were highly correlated with color parameter results obtained from a handheld colorimeter, which is an alternative method for detecting flour quality issues. In addition, the scanner method was also successful in detecting brown specks from bran and distinguishing them from black specks from smut. This is an important alternative application of this method as the presence of bran in flour can indicate sifting problems during the milling process. Altogether, this study shows that flatbed scanners can rapidly detect quality issues from smut and bran contamination, which can help ensure the production of high-quality flour in mills.

Technical Abstract: Flour quality can be evaluated by several methods to detect color changes caused by contamination like fungal damaged kernels, but many existing methods are time-consuming and require specialized training. In this study, a commercial flatbed scanner was used to quickly detect and quantify the abundance of black specks derived from smutty grains in wheat flour samples. This method was developed and validated using wheat samples collected from the field and contained a range of smut contamination. Although specks could easily be detected and counted, we found that speck counts varied with scanner resolution setting. Therefore, an alternate parameter referred to as “%area-smut” was calculated and resulted in a more consistent values per sample regardless of scanner resolution. Our method easily classified flour samples as clean, marginally clean, or contaminated by using varied levels of %area-smut. Clean flour samples were defined as %area-smut was below 0.025% while contaminated flours had %area-smut greater than 0.050%. Marginal flours had %area-smut from 0.025%-0.050%. Additionally, flour color parameters, L*a*b*, were determined for each scanned images using imaging processing software. These image values were well correlated with color values measured with a reference hand-held colorimeter. Moreover, the %area-smut detected by the scanner was found to be inversely related to the flour color brightness parameter (L) and directly correlated with the number of smutty seeds manually detected in a 250 g sample. Therefore, this method represents a rapid and reliable way to distinguish clean flour from flour derived from wheat containing various levels of smut contamination.