20.500.12556/DKUM-22510 AVTOMATSKO PREPOZNAVANJE OBRAZOV V IDEO SIGNALU V REALNEM ČASU REAL TIME FACE RECOGNITION IN VIDEO V diplomski nalogi smo opisali digitalno sliko ter uporabljene postopke procesiranja digitalne slike, postopek iskanja objekta/obraza v digitalni sliki s Haarovim kaskadnim klasifikatorjem, uporabljen postopek detekcije gibanja ter nekaj postopkov prepoznavanja obrazov. Najbolj natančno smo opisali metodo glavnih komponent ali PCA, ki smo jo tudi implementirali. Implementacijo PCA smo združili v aplikacijo, ki implementira avtorizacijo vstopa v laboratorij in prilagajanje višine interaktivne platforme BQ-portala višini uporabnika. Izvedbo algoritmov smo tudi opisali. Nadalje smo izvedli preizkus, kako nekateri dejavniki pri postopku in prepoznavanja obrazov vplivajo na uspešnost prepoznavanja. Na koncu smo še našteli nekatere druge možnost uporabe opisanih tehnologij. In this graduation thesis, we have presented and described the digital image and the used methods for digital image processing, the Haar cascade classifier for face/object detection on digital image, the used algorithms for motion detection and some algorithms used for face recognition. In most detail, we have described the Principal Component Analysis face recognition algorithm (PCA) which we have also implemented. We have also combined our PCA implementation with laboratory entrance authorization and interactive platform of BQ-portal to user-height-adaptation application. In addition, we have described the implementation of all the algorithms. Furthermore, we performed the tests of how some of factors of face learning and face recognition affect the recognition accuracy. Finally, we have listed some of the possibilities of the use of the described technologies. prepoznavanje obrazov v realnem času detekcija obrazov Haarov kaskadni klasifikator detekcija gibanja procesiranje digitalne slike PCA metoda glavnih komponent Real time face recognition Face detection Haar cascade classifier motion detection digital image processing PCA Principal Component Analysis true true false [M. Dimec] Slovenski jezik Angleški jezik Diplomsko delo 2012-03-29 15:20:39 2012-05-10 14:31:30 2022-04-14 11:52:06 2023-12-30 03:24:04 2012 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 681.883.05:004.9(043.2) 16203030 URN:SI:UM:DK:MBTMGZDZ UNI_Dimec_Mihael_2012.pdf UNI_Dimec_Mihael_2012.pdf 1 D2C1985D0A628C8EE424BB30B5F9EE53 753aa469046cc403dffe3bc9db62bd2b149f6528fd29cfabb1ecae712d606032 8b34d490-7c06-11eb-bb7a-00155d0001ca 20.500.12556/dkum/aaccd037-0361-435e-ba2a-4d5ee1543bab https://dk.um.si/Dokument.php?lang=slv&id=29365 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0