Machine Learning and Artificial Intelligence can help a great deal in vineyard operations and management to produce the best wine possible. An article discussing 5 ways that this can be done has been published in the Pursuit magazine from The university of Melbourne that showcase CUTTING-EDGE RESEARCH AND INSIGHTFUL COMMENTARY BY WORLD-LEADING EXPERTS.
Press HERE for the full paper
Whether you are a student, farmer, corporate person, or policy maker, you are most likely aware that the Earth is changing, and not always for the better. Take the rise of the Earth’s global surface temperature since 1880 through today as an example. Our Earth is “redder” and “most of the warming has occurred in the past 35 years” (NASA, 2017).
Five-Year Global Temperature Anomalies from 1880 to 2016 (NASA, 2017)
Whilst global warming is the “big picture” on how the Earth is changing, our day-to-day life is also being affected, in terms of what we consume. For most of us, our food comes from the local market or food vendors. However, its main source is derived from farms that are facing climate-related challenges such as extreme weather events, pests and diseases, which require quick adaptations. And if farms cannot adapt, our food supply chain may be disrupted.
Thus, since 2012, the Vineyard of the Future (VoF), has aimed to establish a fully-instrumented vineyard using the Internet of Things (IoT) such as Unmanned Aerial Vehicles (UAVs), remote sensors, and apps. It has also acted as a test-bed for new technology applications and investigated the potential effects of climate change in different agricultural fields. To date, the VoF has worked on practical solutions such as:
- The free VitiCanopy App that analyses the leaf area index and canopy porosity of grapevines, whereby its parameters can be related to berry quality parameters such as anthocyanin content and polyphenols.
- Precision viticulture using UAVs for data collection on how grapevines are affected by abiotic factors (post-effect) such as frosts (Baofeng et al., 2016).
- Biological sensors (dogs) to detect different compounds of interests, and pests such as phylloxera in viticulture (Fuentes & De Bei, 2016).
- Robotic pourers and computer vision techniques to assess the quality traits of sparkling wines/beers based on foamability and bubble dynamics (Fuentes & De Bei, 2016).
- The BIOSENSORY app that decodes consumer behaviour using facial biometrics, as our physiological response to stimuli is before our verbalisation of it.
In addition, new technology applications are also being developed for specific monitoring, and these include:
- The early (pre-effect) detection on frost damage assessments in vineyards.
- A pilot app for apples to detect sunburn risks and model final fruit-size during harvest time.
- The detection of smoke contamination in vineyards, whereby the smoke-related compound guaiacol glycoconjugates results in undesirable aromas and flavours in wine (Fuentes & Tongson, 2017).
- The use of non-invasive remote sensing to assess meat quality, whereby biometrics such as breathing patterns, body temperature, and heart-rate, are used to quantify the stress levels of cows.
As such, this list provides a snapshot of the VoF’s main projects in viticulture, fruit production, sensory science and animal science. Nevertheless, the end-goal remains, and that is for these solutions to be transferable to all fields of agriculture.
Baofeng, S., Jinru, X., Chunyu, X., Yulin, F., Yuyang, S. and Fuentes, S., 2016. Digital surface model applied to unmanned aerial vehicle based photogrammetry to assess potential biotic or abiotic effects on grapevine canopies. International Journal of Agricultural and Biological Engineering, 9(6), p.119.
Fuentes, S. and De Bei, R., 2016. Innovations and technology: Advances of the Vineyard of the Future initiative in viticultural, sensory science and technology development. Wine & Viticulture Journal, 31(3), p.53.
Fuentes, S. and Tongson, E., 2017. Vinyard technology: Advances in smoke contamination detection systems for grapevine canopies and berries. Wine & Viticulture Journal, 32(3), p.36.
NASA, 2017. Scientific Visualization Studio. [Online] Available at: https://svs.gsfc.nasa.gov/4546 [Accessed 11 December 2017].
2017 June 25th – 28th June. Fourth International Conference on Cocoa, Coffee and Tea (Poster). Turin, Italy.
Chacon G., Fuentes S., Gonzalez Viejo C., Zhang P.
What a robot can tell you about beer quality? Implementing the RoboBEER to assess beer quality based on sensory descriptors
Claudia Gonzalez Viejo 1, Sigfredo Fuentes 1*, Damir Torrico1, Kate Howell1, and Frank R. Dunshea1.
1 University of Melbourne, School of Agriculture and Food, Faculty of Veterinary and Agricultural Sciences, VIC 3010, Australia
* Correspondence: firstname.lastname@example.org; Tel.: +61 3 9035 9670
Journal of The Science of Food and Agriculture
Full Article: CLICK HERE
Beer quality is mainly defined by its color, foamability and foam stability, which are influenced by the chemical composition of the product such as proteins, carbohydrates, pH and alcohol. Traditional methods to assess specific chemical compounds are usually time-consuming and costly. This study used rapid methods to evaluate 15 foam and color-related parameters using a robotic pourer (RoboBEER) and chemical fingerprinting using near infrared spectrometry (NIR) from six replicates of 21 beers from three types of fermentation. Results from NIR were used to create partial least squares regression (PLS) and artificial neural networks (ANN) models to predict four chemometrics such as pH, alcohol, Brix and maximum volume of foam.