SWe WeFS A Sem antically Enabled Fire Monitoring Application K. - - PowerPoint PPT Presentation

▶
swe wefs
SMART_READER_LITE
LIVE PREVIEW

SWe WeFS A Sem antically Enabled Fire Monitoring Application K. - - PowerPoint PPT Presentation

Aug 05, 2023 234 likes •337 views

SWe WeFS A Sem antically Enabled Fire Monitoring Application K. Kyzirakos, M. Karpathiotakis, G. Garbis , C. Nikolaou, K. Bereta, I. Papoutsis, T. Herekakis, D. Michail, M. Koubarakis, C. Kontoes National and Kapodistrian University of Athens

slide-1
SLIDE 1

A Sem antically Enabled Fire Monitoring Application

  • K. Kyzirakos, M. Karpathiotakis, G. Garbis, C. Nikolaou,
  • K. Bereta, I. Papoutsis, T. Herekakis, D. Michail,
  • M. Koubarakis, C. Kontoes

National and Kapodistrian University of Athens National Observatory of Athens Harokopeio University of Athens

Sem antic W eb Challenge I nternational Sem antic W eb Conference ( 2 0 1 2 )

1 4 / 1 1 / 2 0 1 2

SWe WeFS

slide-2
SLIDE 2

Outline

  • The Fire Monitoring Service of the National

Observatory of Athens (NOA) http://papos.space.noa.gr/fend_static

  • Improving the service using semantic technologies
  • Representing fire related information using
  • ntologies
  • Enriching products with linked geospatial data
  • Refining products
  • Evaluation
slide-3
SLIDE 3

MSG-1 Seviri (5 mins) MSG-2 Seviri (15 mins)

Fire m onitoring application

Pre-TELEIOS practice

Disk Array Eumetsat @ 9.5°East

METEOSAT Ground Station Raw data are decoded and stored temporary as wavelet compressed images @...

SEVIRI Monitor

SQLite

PostGIS System users ( Civil Protection Authorities etc.) Manage SEVIRI data stream in real time:

  • Extract & store raw file metadata
  • Filter & dispatch raw MSG products
  • Remotely trigger processing chain
  • Dispatch processed products
slide-4
SLIDE 4

Processing Chain (SciQL based) HotSpots Back End: MonetDB / Strabon

Fire m onitoring application

Advancements – integration of the TELEIOS technologies

4 5/16/2013 Data Vault Eumetsat @ 9.5°East Raw Data

  • Corine Landcover
  • Admin Boundaries
  • POIs

External Sources

  • Search & Display
  • Search for raw & Processing
  • Real-time Fire Monitoring
  • Refinement (Post-Processing)
  • Linked Data

Geospatial Ontology

Web access based on Semantics Linked Geospatial Data Semantic technologies

slide-5
SLIDE 5

Linked Geospatial Data

  • Datasets that we developed and published as

linked data:

  • Corine Land Use / Land Cover
  • Coastline of Greece
  • Greek Administrative Geography
  • Portal: http:/ / w w w .linkedopendata.gr/
  • Datasets from Linked Open Data Cloud
  • LinkedGeoData
  • GeoNames
slide-6
SLIDE 6

Sem antic Enrichm ent for Hotspots

  • Enrich hotspot products
  • 1. Connect each hotspot with a

municipality that it is located

  • I m prove accuracy with

respect to underlying area

  • 2. Eliminate false alarms in sea
  • 3. Eliminate false alarms in

inconsistent land cover areas

  • 4. Keep land part of the

polygon

  • I m prove accuracy with

respect to tem poral persistence of each hotspots

5. Remove “Christmas tree” effects

”Christm as tree effect”: some hotspots appear in a timestamp, in the next timestamp they disappear, then they re-appear again, and so on.

slide-7
SLIDE 7

Evaluation of perform ance

1 10 100 1000 10000 Response time (msec, logscale)

Satellite image acquisitions (timestamp)

Store Municipalities Delete in the sea Other inconsistent areas Refine in coast Time persistence

slide-8
SLIDE 8
  • Fully automatic while the previous service had

many manual steps

  • Excellent performance
  • Fully operational and on the Web
  • Used on a daily basis during and after the

events in the summer of 2012 by

  • Greek Civil Protection Agency
  • Greek Fire Brigade
  • Greek Army
  • Initial user comments very encouraging

9

W hy is it im portant?

slide-9
SLIDE 9
  • We developed a new geospatial RDF store

(Strabon)

  • We developed a new query language for

linked geospatial data (stSPARQL)

  • We published new linked geospatial datasets

(some of the very few available ones with complex geometries)

  • The stSPARQL update statements are very

complex (many triple patterns, many SPARQL constructs)

1 0

W hy has it been a challenge?

slide-10
SLIDE 10

Any questions?

  • Strabon
  • Manolis Koubarakis, Kostis Kyzirakos, Manos Karpathiotakis,

Charalampos Nikolaou, Giorgos Garbis, Konstantina Bereta, Kallirroi Dogani , Stella Giannakopoulou and Panayiotis Smeros.

  • Web site: http://strabon.di.uoa.gr
  • Mercurial repository: http://hg.strabon.di.uoa.gr
  • Trac: http://bug.strabon.di.uoa.gr
  • Mailing list: http://cgi.di.uoa.gr/~mailman/listinfo/strabon-users
  • Real Time Fire Monitoring Service
  • http://papos.space.noa.gr/fend_static
  • Greek Linked Open Data
  • http://www.linkedopendata.gr
  • TELEIOS EU Project
  • http://www.earthobservatory.eu

1 1

Thank you for you attentionn!

SWe WeFS