Pregled bibliografske jedinice broj: 3692
Computer Vision System for Automatic Tracking Coloured Objects
Computer Vision System for Automatic Tracking Coloured Objects // Proceedings of the 20th International Convention : MIPRO '97 / Biljanović, Petar ; Skala, Karolj ; Ribarić, Slobodan ; Budin, L (ur.).
Rijeka: Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO, 1997. str. 177-180 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 3692 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Computer Vision System for Automatic Tracking Coloured Objects
Autori
Ribarić, Slobodan ; Gluhak, Denis
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the 20th International Convention : MIPRO '97
/ Biljanović, Petar ; Skala, Karolj ; Ribarić, Slobodan ; Budin, L - Rijeka : Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO, 1997, 177-180
ISBN
953-6042-40-1
Skup
International Convention MIPRO (20 ; 1997)
Mjesto i datum
Opatija, Hrvatska, 19.05.1997. - 23.05.1997
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
computer vision; dynamic vision; object tracking; colour
Sažetak
A computer vision system for automatic tracking of multiple moving coloured objects on different textures is described in the paper. The extracting of objects from background is performed in the peripheral phase. The basic feature for extracting an object from background is its colour. An object may be coloured by one of seven different colours which are represented as corners of the RGB colour cube. The position of the moving object at particular instant in time is determined in the peripheral phase. Objects are recognized by using the generalized perceptron algorithm in the attention phase. The system generates trajectories of moving objects in the real-time and calculates attributes which are needed for a reconstruction and analysis of objects moving. This information can be effectively used during the cognitive phase of a dynamic computer vision system.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb
Profili:
Slobodan Ribarić
(autor)