Title :
Automatic threshold selection for automated visual surveillance
Author :
Çelik, Turgay ; Kabakli, Tarik ; Uyguroklu, M. ; Özkaramanli, Huseyin ; Demirel, Hasan
Author_Institution :
Ileri Teknoloji Arastirma ve Gelistirme Enstitusu, Dogu Akdeniz Universitesi, Gazimagusa, Turkey
Abstract :
Automated visual surveillance systems mostly depend on an effective background subtraction technique. Most background subtraction techniques suffer mainly from parameter updates for threshold selection. A new threshold selection technique, which is found while training the system to learn the background, is proposed.
Keywords :
learning (artificial intelligence); surveillance; video signal processing; automated visual surveillance; automatic threshold selection; background subtraction technique; parameter updates; training; Parametric statistics; Subtraction techniques; Surveillance;
Conference_Titel :
Signal Processing and Communications Applications Conference, 2004. Proceedings of the IEEE 12th
Print_ISBN :
0-7803-8318-4
DOI :
10.1109/SIU.2004.1338568
