Botanical origin identification of Sicilian honeys based on artificial senses and multi-sensor data fusion

Quality assessment of honey is often related to its botanical origin, which is traditionally evaluated by a melissopalynology analysis. This technique is quite laborious, time-consuming and requires a high-skilled and trained technician. Another major limitation of this method is fraud addition of pollen for authenticity. Conversely, artificial senses, coupled with chemometrics, have been shown to be rapid and reliable tools for the discrimination of honeys botanical origin. Further, the combined use of these techniques should improve sample classification; however, the main challenge is how to combine data coming from each device. In this study, a comparison between data from single modality and fusion methods, to classify different Sicilian honey varieties, was presented. Combining a potentiometric electronic tongue and a computer vision system, a satisfying recognizing percentage was achieved. This approach proved to be fast, simple and inexpensive; furthermore, any sample preparation or the use of chemicals was required.