BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.staging.osgeo.org//foss4g-2022//speaker//KZNCCL
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-RGYMR3@talks.staging.osgeo.org
DTSTART;TZID=CET:20220824T171500
DTEND;TZID=CET:20220824T174500
DESCRIPTION:High resolution aerial photos combined with accurate map data r
 epresents a perfect data set for training artificial intelligence models. 
 The ‘KartAi’ project is an innovation project in public sector aimed a
 t developing Ai-methods that detects buildings not in the cadastre or the 
 building map dataset. Thereafter involving the property owner/citizen in a
  digital dialog and validate or crowdsource more detailed data. The founda
 tion for this is high quality datasets for training and validating the dif
 ferent Ai-models. High resolution aerial photos are collected in large par
 ts of Norway on a regular basis – often yearly – in a collaboration be
 tween federal and municipal. Thereby there exists a vast amount of extreme
 ly detailed image data combined with building map data and cadastre data. 
 However\, training the Ai-models have uncovered that minor errors and ‘s
 kewed’ photos and/or vector data affects the results of the segmentation
  of roof tops/buildings. Therefore the KartAi projects has made fine tuned
  and accurate training data sets in several geographical areas optimized f
 or training on detecting and segmenting buildings. \nIn several large scal
 e experiments\, a multitude of existing models\, newer models and own mode
 ls have been training and validated. Additionally we have included LIDAR-h
 eight data to enhance the precision of segmenting between the likes of roo
 fs and terraces. Training the models on the existing data yields good resu
 lts. However\, when finetuning with the high accurate data – the models 
 show impressing results. \nSpatial Ai projects like KartAi are at the merc
 y of volumes of good training data. Our experience show that even more acc
 urate data sets improve the models even further. Therefore\, the project h
 as made efforts that have resulted in the release of the training data set
 s publicly – as well as all of the results data for the different models
  and approaches that have been developed. This is an effort into developin
 g a more open living lab for Spatial Ai in Norway. Our hope is that sharin
 g the knowledge and data created can ensure that other Ai-models have easi
 er access to high resolution and high accuracy data – to train models in
  the open living lab – and apply the models internationally where data i
 s scarcer.
DTSTAMP:20260404T094514Z
LOCATION:Auditorium
SUMMARY:KartAi – An open living lab for Ai in Norway - Alexander Salveson
  Nossum
URL:https://talks.staging.osgeo.org/foss4g-2022/talk/RGYMR3/
END:VEVENT
BEGIN:VEVENT
UID:pretalx-foss4g-2022-ESLMWH@talks.staging.osgeo.org
DTSTART;TZID=CET:20220826T101500
DTEND;TZID=CET:20220826T102000
DESCRIPTION:Sensordata (IoT) is widespread in both private and public secto
 r. However\, making use of sensordata across different sectors and applica
 tions is challenging - in particular with respect to a geospatial applicat
 ion across different use cases. This encompasses both enviroment/climate s
 ensors\, like water-level sensors to smart-building monitoring and water p
 ipe sensors. An interdisciplinary team from diverse sectors is working tow
 ards building national standards\, an open architecture and implementing p
 roof-of-concepts on a national sensorhub for sharing streams and archives 
 of sensordata in Norway. The team builds upon the very successful open dat
 a ecosystems (SDI) that exists in Norway for standardized geospatial data.
  The project is funded from a range of partners including municipalities\,
  the mapping authority and the maritime ports of Norway. The working group
  includes open source tech expertise on sensor technology alongside user a
 nd demand expertise from the different sectors. \n\nThis talk will focus o
 n the technological advances made from the team both on software and archi
 tecture. There will be particular focus on the open architecture and softw
 are prototyping that has been developed in the working group. Both of whic
 h will be available under an open license.
DTSTAMP:20260404T094514Z
LOCATION:Room 4
SUMMARY:Norwegian National SensorHub - sharing IoT data with open standards
  and technology - Alexander Salveson Nossum
URL:https://talks.staging.osgeo.org/foss4g-2022/talk/ESLMWH/
END:VEVENT
END:VCALENDAR