crta
Hrvatska znanstvena Sekcija img
bibliografija
3 gif
 Naslovna
 O projektu
 FAQ
 Kontakt
4 gif
Pregledavanje radova
Jednostavno pretraživanje
Napredno pretraživanje
Skupni podaci
Upis novih radova
Upute
Ispravci prijavljenih radova
Ostale bibliografije
Slični projekti
 Bibliografske baze podataka

Pregled bibliografske jedinice broj: 764238

Zbornik radova

Autori: Bugarić, Marin; Stipaničev, Darko; Šerić, Ljiljana; Jakovčević, Toni; Krstinić, Damir
Naslov: Fusion of wildfire risk index and automatic video based wildfire surveillance systems
Izvornik: II International Conference on Fire Behaviour and Risk - Book of Abstracts
Dio CC časopisa: NE
Skup: II International Conference on Fire Behaviour and Risk
Mjesto i datum: Alghero, Italija, 26-29.05.2015.
Ključne riječi: wildfires; wildfire risk index; Adriatic region
Sažetak:
Determining the probability of fire occurrence and fire spread at certain area and time could be useful for all fire-fighting activities. A numerical measure used to describe this probability is usually referred to as a wildfire risk index, and as such is traditionally used to enhance human vigilance to wildfires in the pre-fire phase. Higher values of wildfire risk index require elevated level of alertness and increased concentration not only for fire-fighters, but also for observers responsible for the early detection of wildfire. In this paper, our focus is primarily on the application of wildfire risk index estimation in the early detection of wildfires. Two types of wildfire surveillance co-exist today: the traditional human based wildfire surveillance and a more advanced video based wildfire surveillance with implemented automatic fire and smoke detection. These advanced video based systems are increasingly evolving and applied in more and more regions affected by wildfires. The importance of wildfire risk index estimation in traditional human based wildfire surveillance is quite well known - the wildfire risk index estimation is generally used to increase the wildfire observer’s concentration. On the other hand, wildfire risk index estimation is still not widely used in automatic video based wildfire surveillance systems. Because of this, this paper focuses more on the following topic: How can the wildfire risk index be used to improve automatic video-based wildfire surveillance systems? To successfully introduce wildfire risk index into automatic wildfire surveillance, the estimated index should have as high spatial resolution as possible. General, low-resolution wildfire risk indexes calculated in most of the countries exposed to wildfires are not adequate for this purpose. The wildfire risk index should be calculated on a micro location scale. One example of such index is the Site-Specific Wildfire Risk Index (SWRI). In the first part of this paper we shortly introduce the method of calculating the Site- specific Wildfire Risk Index (SWRI). In the following, we specify how this Site-Specific Wildfire Risk Index maps could be used in automatic video based wildfire surveillance systems. We primarily focus on the problem that is equivalent to human based wildfire surveillance - how to increase (or decrease) the wildfire observer’s concentration, but in this case - how to increase (or decrease) the automatic video based wildfire surveillance detection sensitivity. Detection quality of most wildfire surveillance systems depends largely on chosen sensitivity parameters. High sensitivity may result in a large number of false alarms, while a low sensitivity may result in a missed detection. Human observers rarely change those sensitivity parameters manually. It can be shown that wildfire risk information can be used to dynamically change those parameters in order to achieve a better detection. Various approaches are discussed, such as the sensitivity adjustment based on the average risk index in the visible image, as well as the more complicated one based on Augmented Reality (AR). These topics are the subject of research of our Department in the last couple of years.
Vrsta sudjelovanja: Predavanje
Vrsta prezentacije u zborniku: Sažetak
Vrsta recenzije: Međunarodna recenzija
Izvorni jezik: ENG
Kategorija: Znanstveni
Znanstvena područja:
Računarstvo
Tiskani medij: da
Upisao u CROSBI: Marin Bugarić (Marin.Bugaric@fesb.hr), 5. Lip. 2015. u 09:54 sati



Verzija za printanje   za tiskati


upomoc
foot_4