VeriVin’s technology will pave the way for the discrimination of counterfeit wine and spirits, in bottle, with no need to take the cork out. This involves a multi-component analysis to screen for anomalies which could indicate that the liquid in the bottle is not what is stated on the label.
A simple principal components model for a given wine can already give an accurate mathematical description of a bottle of wine or spirits. If many bottles of the same product are tested, it is possible to spot those that are different from the others and classify them as outliers. These are ‘suspect bottles’ that are most likely faulty, fake or adulterated (which can in turn be deduced by looking at the variation in the principal components). We can make these differentiations increasingly specific by using advanced machine learning techniques.
Figure 1 shows the results of an experiment in which we were able to successfully single out heavily oxidised bottles from a single batch of the same wine (wine A) and classify two different wines (wine A and wine B) according to their spectral response.
Aside from verifying authenticity, our technology could also be used to verify compliance with regulations. Adulteration of wine and spirits can take on many forms. The addition of sugars, acids, water (dilution), volatile oils (aromas), synthetic sweeteners, juices of other fruits, and other synthetic substances (which in the past have included toxic ones) are all possibilities.
Another possibility is the use of additives in doses larger than those allowed by the regulations governing the particular winemaking area of interest. In well established wine producing regions like the EU, laboratory analyses to screen for possible adulteration are required by law and are routinely carried out, but this occurs prior to bottling. Our technology will enable the end consumer or supplier to test for adulteration in-situ, and in a non-invasive way.
Back to Our Technology & Applications.