BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211207T055401Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20211118T083000
DTEND;TZID=America/Chicago:20211118T170000
UID:submissions.supercomputing.org_SC21_sess280_rpost101@linklings.com
SUMMARY:Monitoring Urban Changes with Ensemble of Neural Networks and Deep
 -Temporal Remote Sensing Data
DESCRIPTION:Posters, Research Posters\n\nMonitoring Urban Changes with Ens
 emble of Neural Networks and Deep-Temporal Remote Sensing Data\n\nZitzlsbe
 rger, Podhoranyi, Svatoň, Lazecký, Martinovič\n\nUrban change de
 tection with remote sensing data covers a wide field of applications like 
 understanding socio-economic impacts, identifying new settlements, or anal
 yzing trends of urban sprawl. It is used for decades. Analyses, however, a
 re usually carried out manually by selecting high-quality samples, restric
 ted to small scale scenarios either temporarily limited or with low spatia
 l or temporal resolution. To process a large amount of available remote se
 nsing observations for a selected period, we propose a fully automated met
 hod to train an ensemble of neural networks, without the need to manually 
 select samples. We consider two eras with three sites, each with at least 
 500 km^2, and deep observation time series with hundreds up to over a thou
 sand combined synthetic aperture radar (SAR) and multispectral optical obs
 ervations. In order to train such large data sets, we apply data-parallel 
 deep learning with Horovod and multiple NVIDIA Tesla V100 GPUs.\n\nRegistr
 ation Category: Tech Program Reg Pass, Exhibit Hall Only
END:VEVENT
END:VCALENDAR