Pregled bibliografske jedinice broj: 240822
Cluster Image Processing in Ceramic Tiles Quality Evaluation
Cluster Image Processing in Ceramic Tiles Quality Evaluation // Proceedings of 29th International Conference MIPRO 2006, HGS-Hypermedia and Grid Systems / Skala, Karolj (ur.).
Rijeka: Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO, 2006. str. - (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 240822 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Cluster Image Processing in Ceramic Tiles Quality Evaluation
Autori
Vasilić, Suzana ; Hocenski, Željko ; Hocenski, Verica
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of 29th International Conference MIPRO 2006, HGS-Hypermedia and Grid Systems
/ Skala, Karolj - Rijeka : Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO, 2006
Skup
29th International Conference MIPRO 2006
Mjesto i datum
Opatija, Hrvatska, 22.05.2006. - 26.05.2006
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
image processing; cluster; ceramic tiles; quality
Sažetak
Clusters are grid basic elements consisted of the independent computers working as one unique computer for realization of greater computer power. They are often used for execution of demanding scientific and technological calculations. The CRO-GRID is national grid, composed of five connected geographically distanced clusters. ETFOS Cluster is one of the CRO-GRID's clusters. It is Beowulf type of cluster with one central node. In this work it is used for processing of RGB images of ceramic tiles. The pictures are converted into binary pictures using global threshold. We determine border of an object in the picture, starting by determining the size of the object. In order to determine the size, we look for the most prominent pixel on the every picture side: left, right, up and down. After this, every edge is divided on the given number of segments and we look for marginal pixel of every segment. Values of marginal pixels are saved in vectors in order to analyze the orientation of pictures. The processing is performed on personal computer and on the cluster for comparison of necessary processing run times.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo