BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.staging.osgeo.org//foss4g-2022//speaker//BBMTUR
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-2022-KL8DQL@talks.staging.osgeo.org
DTSTART;TZID=CET:20220826T100000
DTEND;TZID=CET:20220826T103000
DESCRIPTION:We discuss the "Diet Hadrade" codebases\, which provides an ope
 n-source\, lightweight mechanism for leveraging remote sensing imagery and
  machine learning techniques to aid in humanitarian assistance and disaste
 r response (HADR) in austere environments. \n\nIn a disaster scenario (be 
 it an earthquake or an invasion) where communications are unreliable\, ove
 rhead imagery often provides the first glimpse into what is happening on t
 he ground. The rapid identification of both vehicles and road networks dir
 ectly from overhead imagery allows a host of problems to be tackled\, such
  as congestion mitigation\, optimized logistics\, evacuation routing\, etc
 .  Such challenges often arise in the aftermath of natural disasters\, but
  are also present in crises like the current invasion of Ukraine where roa
 ds are choked with civilians fleeing the fighting.\n\nAutomobiles provide 
 an attactive proxy for human popuplation due to their mobile nature and th
 e necessity of population movement in many disaster scenarios.  In this pr
 oject\, we deploy the YOLTv5 computer vision object detection codebase to 
 rapidly identify and geolocate vehicles over large areas. Vehicle detectio
 ns yield significantly greater utility when combined with road network dat
 a. We use the CRESI computer vision framework to extract up-to-date road n
 etworks with travel time estimates\, thus permitting optimized routing. Th
 e CRESI codebase is able to extract roads using only overhead imagery\, so
  flooded areas or obstructed roadways will sever the CRESI road graph\; th
 is is crucial for post-disaster scenarios where existing road maps may be 
 out of date and the route suggested by cloud navigation services may be im
 passable or hazardous.\n\nDiet Hadrade provides a number of graph theory a
 nalytics that combine the CRESI road graph with YOLTv5 locations of vehicl
 es. We combine the car detections with the road network to infer how conge
 sted certain areas are. Congestion information is important for everyday l
 ife\, but also crucially important in disaster response scenarios when roa
 ds may become impassable due to both natural phenomena as well as traffic.
   \n\nWe leverage the detailed road graph and vehicle location information
  to illustrate a number of scenarios\, such as: bulk evacutation\, optimal
  aid disbursement locations\, critical intersections\, and detection and a
 utomated avoidance of dangerous locales.  These capabilities are presented
  in an interactive dashboard that computes optimal routes on the fly based
  on user inputs.
DTSTAMP:20260411T101832Z
LOCATION:General online
SUMMARY:The Potent Mix of Computer Vision\, Graph Theory\, Satellite Imager
 y\, Vehicles\, and Roads - Adam Van Etten
URL:https://talks.staging.osgeo.org/foss4g-2022/talk/KL8DQL/
END:VEVENT
END:VCALENDAR