BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Europe/Lisbon
BEGIN:DAYLIGHT
DTSTART:20210328T020000
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:WEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20201025T010000
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:WET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220317T225136Z
UID:F298C31C-0109-4413-84B9-B5CC4251B7C8
DTSTART;TZID=Europe/Lisbon:20210224T170000
DTEND;TZID=Europe/Lisbon:20210224T180000
DESCRIPTION:The wine industry has been striving to achieve differentiation 
 in their products and to improve their quality and consistency\, which inv
 olves harvesting grapes at the optimal maturity point and selecting them a
 ccording to the desired characteristics of the wine to be produced. In thi
 s context\, hyperspectral imaging (HSI) combined with machine learning alg
 orithms (ML) is a promising alternative to predict important oenological p
 arameters and assist on harvesting critical decisions. However\, the large
  amount of data generated by HSI\, together with the large variability ass
 ociated with the problem (varieties involved\, climate\, terroir\, etc.)\,
  raise unusual challenges for data-driven modelling. Several ML approaches
  have been proposed to handle such data characteristics\, but selecting a 
 suitable methodology that best address the problem under study and make su
 re it generalizes well\, is a cumbersome task. Our work is focused in two 
 fundamental and novel aspects to address the natural variability arising f
 rom different grape varieties\, vintages and growth conditions: the essent
 ial wavelength bands selection (with the purpose of reducing the dimension
 ality of data without losing predictive power) and the generalization abil
 ity of the ML model under such demanding conditions.\n\nVirtual: https://e
 vents.vtools.ieee.org/m/262310
LOCATION:Virtual: https://events.vtools.ieee.org/m/262310
ORGANIZER:jmac@ieee.org
SEQUENCE:6
SUMMARY:CI Lectures Series - Wine grape ripeness assessment using Hyperspec
 tral Imaging and Machine Learning
URL;VALUE=URI:https://events.vtools.ieee.org/m/262310
X-ALT-DESC:Description: <br /><h6 style="text-align: justify\;">The wine in
 dustry has been striving to achieve differentiation in their products and 
 to improve their quality and consistency\, which involves harvesting grape
 s at the optimal maturity point and selecting them according to the desire
 d characteristics of the wine to be produced. In this context\, hyperspect
 ral imaging (HSI) combined with machine learning algorithms (ML) is a prom
 ising alternative to predict important oenological parameters and assist o
 n harvesting critical decisions. However\, the large amount of data genera
 ted by HSI\, together with the large variability associated with the probl
 em (varieties involved\, climate\, terroir\, etc.)\, raise unusual challen
 ges for data-driven modelling. Several ML approaches have been proposed to
  handle such data characteristics\, but selecting a suitable methodology t
 hat best address the problem under study and make sure it generalizes well
 \, is a cumbersome task. Our work is focused in two fundamental and novel 
 aspects to address the natural variability arising from different grape va
 rieties\, vintages and growth conditions: the essential wavelength bands s
 election (with the purpose of reducing the dimensionality of data without 
 losing predictive power) and the generalization ability of the ML model un
 der such demanding conditions.</h6>
END:VEVENT
END:VCALENDAR