Welcome to the Wildfire Observers and Smoke Recognition Homepage
About the smoke recognition portal PDF Print
Monday, 08 February 2010 13:35

Forest fires and generally wildfires represent a constant threat to ecological systems, infrastructure and human lives. In the last decade the threat of forest fires has increased significantly because of the climate changes. Apart from preventive measures, early fire detection on one side and quick and appropriate intervention on the other, are measures of vital importance for wildfire damage minimization.

Early fire detection is traditionally based on human wildfire surveillance, but modern Information-Communication Technologies (ICT) could offer advanced automatic wildfire observers as a replacement for human wildfire observation.  Modern automatic wildfire observers could be used not only for early fire detection, but also for distant video presence, the feature quite important in fire fighting stage.

This Web portal is dedicated to automatic wildfire observers and particularly focused on the automatic detection and recognition of the smoke generated by wildfire, because the smoke detection is still the most important feature of today’s automatic wildfire observers.

Wildfire observers and smoke recognition portal is intended to be a place for collaboration regarding every aspect of automated smoke-detection process. That includes different methods and algorithms for wildfire smoke-detection, image and video databases, definition of metrics regarding algorithm evaluation and similar. It is planned to be a collaborative site, where different engineers contribute their experience trough writing articles, adding smoke images for testing and training, and of course developing different algorithms for smoke detection.

The Web site is established and maintained by Center for wildfire research which belongs to University of Split Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture.


Last Updated on Saturday, 06 November 2010 08:29
Wildfire Surveillance and Monitoring PDF Print
Written by Darko Stipanicev   
Saturday, 06 November 2010 08:29

Four types of wildfire surveillance and monitoring could be distinguished:

  • Traditional human wildfire surveillance and monitoring.
  • Video cameras based human wildfire surveillance and monitoring.
  • Automatic wildfire detection.
  • Advance automatic wildfire surveillance and monitoring.
Last Updated on Friday, 01 December 2023 18:03
Detection quality metrics PDF Print
Written by Toni Jakovcevic   
Wednesday, 10 February 2010 12:45

Although many video based smoke-detection algorithms have been developed and applied in various experimental or real life applications, the standard method for evaluating their quality has not yet been proposed. Such methodology suitable for smoke-detection algorithms testing and evaluation is presented.

Last Updated on Saturday, 06 November 2010 09:36
Smoke detection algorithms PDF Print
Written by Toni Jakovcevic   
Tuesday, 09 February 2010 13:52

Smoke-detection is a complex task and most of the algorithm authors have their own specific approach to solving such problem. Most of the algorithms have multiple phases or methods of detection (e.g. motion detection followed by region energy estimation) so the different phases of detection are categorised. Some of main categories are specified here, and for each category methodologies from different authors are referenced.

Last Updated on Wednesday, 05 February 2014 20:09
Image and video databases PDF Print
Written by Toni Jakovcevic, Damir Krstinic   
Friday, 05 November 2010 17:33

Very important part in algorithm evaluation is the quality of testing images and/or videos. Larger number of images and/or videos, containing different scenarios and surroundings gives the better quality of the evaluation process.

Last Updated on Friday, 12 September 2014 12:09
Wildfire Observers and Smoke Recognition Homepage, Powered by Joomla!; Joomla templates by SG web hosting