Abstract
This paper aims to develop a machine learning-based system for automaticallydetecting honey adulteration with sugar syrup, based on honey hyperspectralimaging data. First, the floral source of a honey sample is classified by abotanical origin identification subsystem. Then, the sugar syrup adulterationis identified, and its concentration is quantified by an adulteration detectionsubsystem. Both subsystems consist of two steps. The first step involvesextracting relevant features from the honey sample using Linear DiscriminantAnalysis (LDA). In the second step, we utilize the K-Nearest Neighbors (KNN)model to classify the honey botanical origin in the first subsystem andidentify the adulteration level in the second subsystem. We assess the proposedsystem performance on a public honey hyperspectral image dataset. The resultindicates that the proposed system can detect adulteration in honey with anoverall cross-validation accuracy of 96.39%, making it an appropriatealternative to the current chemical-based detection methods.