Published in Scientific Papers. Series D. Animal Science, Vol. LXVIII, Issue 2
Written by Eleonora NEDELCHEVA, Tsvetelina GEORGIEVA, Stanislav PENCHEV, Atanas ATANASOV, Ivaylo HRISTAKOV, Magdalena KACHEL, Plamen DASKALOV
Statistical analysis for evaluation of nine quality parameters of honey samples using two optical devices is presented in the paper. Honey from 14 regions of Bulgaria formed a set of 29 samples included in the study. The presence of heavy metals - arsenic (As), cadmium (Cd), lead (Pb), iron (Fe), Hg and pH, amount of pH, reducing sugars, sweet disaccharide and water content are preliminary evaluated in the certificated laboratory. The color digital images of the honey samples are obtained using camera of mobile phone and document camera and the data are transformed into Lab and HSV color spaces, the feature vector includes 9 color features. Correlation matrices, descriptive statistics and histograms are included in the statistical analysis. Using of document camera allow to apply more informative color features that could increase the assessment accuracy. The results show that there are informative features for quality assessment of honey samples using their color images. The identified informative signs will be integrated into developed mathematical and neural models for the quality assessment of honey through their color images.
[Read full article] [Citation]
