Napredna pretraga

Pregled bibliografske jedinice broj: 910418

Face Template Matching for Ink-Jet Printed Biscuit Tiles


Matić, Tomislav; Vidović, Ivan; Hocenski, Željko
Face Template Matching for Ink-Jet Printed Biscuit Tiles // PROCEEDINGS OF International Conference on Smart Systems and Technologies 2017(SST 2017) / Žagar, Drago ; Martinović, Goran ; Rimac Drlje, Snježana ; Miličević, Kruno (ur.).
Osijek: Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2017. str. 239-244 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)


Naslov
Face Template Matching for Ink-Jet Printed Biscuit Tiles

Autori
Matić, Tomislav ; Vidović, Ivan ; Hocenski, Željko

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
PROCEEDINGS OF International Conference on Smart Systems and Technologies 2017(SST 2017) / Žagar, Drago ; Martinović, Goran ; Rimac Drlje, Snježana ; Miličević, Kruno - Osijek : Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2017, 239-244

ISBN
978-1-5386-3776-0

Skup
International Conference on Smart Systems and Technologies 2017(SST 2017)

Mjesto i datum
Osijek, Hrvatska, 18-20.10.2017

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Ceramic tile ; template matching ; ink-jet printing ; SIFT

Sažetak
One of the recent advances in ceramic tile industrial production is the introduction of ink-jet printers. Ink-jet printers are used for face texture printing in the biscuit tile production stage. This enables production of various ceramic tiles. Different textures and images can be used as the face of the tile. Because of this advance in automatic tile production, it is possible to introduce new methods for biscuit tile face matching in order to improve visual inspection process. In this paper we propose a new approach for biscuit tile face matching. Testing of the proposed method is done on real biscuit tile images acquired on a industrial production line. Presented approach gives 100% classification accuracy for all 60 test images with 12 face images. Future work will include real-time implementation and testing on a production line.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo



POVEZANOST RADA


Ustanove
Fakultet elektrotehnike, računarstva i informacijskih tehnologija Osijek