Pregled bibliografske jedinice broj: 234855
Using Pixel Pairs Difference for Visual Inspection of Ceramic Tiles
Using Pixel Pairs Difference for Visual Inspection of Ceramic Tiles // Tehnički vjesnik / Technical Gazette, 12 (2005), 3-4; 3-9 (međunarodna recenzija, članak, znanstveni)
CROSBI ID: 234855 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Using Pixel Pairs Difference for Visual Inspection of Ceramic Tiles
Autori
Novak, Igor ; Hocenski, Željko ; Slišković, Dražen
Izvornik
Tehnički vjesnik / Technical Gazette (1330-3651) 12
(2005), 3-4;
3-9
Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni
Ključne riječi
ceramic tiles; defect detection; feature extraction; machine vision; pixel pairs difference
Sažetak
The texture feature extraction method could be used in ceramic tiles machine visual inspection. The ordinary method of visual inspection in ceramic tiles production process is performed by a human vision. The results are inhomogeneous tile classes and rather slow and boring inspection. There are papers which describe methods developed for quality inspection of ceramic tiles but these demand a lot of computing power or do not achieve acceptable efficiency. For testing purposes correct and defective ceramic tiles of complex texture are scanned. A number of techniques recently developed to detect defects in textured tiles were reviewed. Results described in this paper show that the approach which combines pixel pairs difference (PPD) with the first order statistical coefficients could be efficiently used in visual defect detection system. PPD method is rather simple and it makes classification based on easy acquirable grayscale images and does not demand significant computing power.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo, Temeljne tehničke znanosti
Citiraj ovu publikaciju:
Časopis indeksira:
- Scopus
Uključenost u ostale bibliografske baze podataka::
- The INSPEC Science Abstracts series
- Cambridge Scientific Abstracts
- Engineering Materials Abstracts
- Corrosion Abstracts
- FLUIDEX
- Elsevier Biobase
- Elsevier GeoAbstracts
- PaperChem