Sensor Web Enabling the Advanced Fire Information System

1
Sensor Web Enabling the Advanced Fire Information
System

• Presented at ISESS 2005
• By Gavin Fleming

2
The Sensor Web

• The Sensor Web is about adding senses to the
  Internet, as well as the intelligence to make
  sense of them
• Open Geospatial Consortium Sensor Web Enablement
  (OGC SWE)
• Making sense of information available through
  Sensor Web
• Sensor Web as enabler of GEOSS, GMES

3
(No Transcript)
4
(No Transcript)
5
AFIS Sensor Web- enablement testbed!

• AFIS v.1 already a Sensor Web-like system with
  proven benefits
• AFIS ideal to help us build SWE architecture in
  agile fashion
• Design
• Implement
• Iterate

6
AFIS Stakeholder Requirements

• Near real time detection of active fires
• Continuous uptime, automated
• Easily accessible from anywhere in the world
• Relevant fire info such as fire location,
  intensity, size, uncertainty
• Web maps e-mail and SMS notifications
• Adaptable alert conditions OGC Alert

7
AFIS stakeholders

• ESKOM
• National and Regional Disaster Management Centres
• Fire Protection Agencies and Brigades
• Forest plantation companies
• National Parks Board
• Western Cape Nature Conservation
• Dept of Water Affairs and Forestry
• SAfNet

8
Flashover 25 July 2004 Highveld region
9
Polar-orbiting Satellite System
Polar Orbiting Satellite System
Scan width 2400 km Overpass times 1000
am 1400 pm 2200 pm 0200 am 1 km
resolution Detection size 0.2 Ht
Scan width 2400 km Overpass times 1000 am -
T 1400 pm - A 2200 pm - T 0200 am -
A Detection size 0.2 ha
10
Geostationary Satellite System
Continues scanning over Europe and Africa Imagery
every 15 minutes 3,5 km resolution Detection
size 5 10 ha
11
AFIS v.1
12
AFIS v.2Sensor WebEnabled
13
Benefits of SWE to AFIS

• Pluggable data sources
• At-(proxy) Sensor processing
• Hybrid service/agent-oriented architecture
  programming to standard interfaces
• Pluggable fire detection algorithms
• Pluggable fire alert agent
• Intelligent agents for other use cases
• All Open Source
• Interoperable, standards based
• Ease of maintenance and collaboration

14
Notifications
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
Plans

• Improve fire detection algorithms
• Implement these as SWE agents
• Make more sensors available
• Weather Humidity Fuel load Wind DEM
• Build semantic query interface
• User-defined alert conditions
• User-defined notification configuration
• Predict, then validate burnt areas

20
(No Transcript)
21
Some Challenges

• Registries, catalogs, metadata, ontologies,
  workflows
• Semantic interoperability
• Working with a small team
• Focus
• Collaborate
• http//wamis.co.za/eskom/checkboxes/sa.htm

22
ACTIVE FIRES 2001 animation 1km MODIS active
fire detections (red) superimposed on MODIS 16
day NDVI