20.500.12556/RUL-147663 Vgrajena zasebnost pri videonadzoru prometa Privacy-By-Design Constrained Traffic Video Surveillance V tej diplomski nalogi je opisan postopek izdelave baze označenih zamegljenih slik cestnega prometa, ki je zasnovana za prihodnje treniranje algoritmov za detekcijo prometa z vgrajeno zasebnostjo. Zameglitev slik je namenjena prikrivanju identitete posameznikov in drugih identifikacijskih podatkov, kar zagotavlja varovanje zasebnosti. Postopek vključuje izdelavo sistema za snemanje prometa z uporabo dveh kamer za Raspberry Pi, ki sta simultano snemali promet – ena je zagotavljala zamegljene in druga ostre posnetke. Slike smo nato poravnali s pomočjo algoritma SIFT. Jasne slike smo označili s pomočjo algoritma DETR, pri čemer so oznake enake tudi za zamegljene slike, saj smo snemanje opravljali z dvema kamerama. Končni rezultat je baza več kot tisoč zamegljenih slik z označbami v formatu COCO. Ta baza je pripravljena za uporabo v nadaljnjih raziskavah in razvoju algoritmov za detekcijo prometa z vgrajeno zasebnostjo. This thesis describes the process of creating a dataset of labeled blurred traffic images, designed for the future training of traffic detection algorithms with built-in privacy. The blurring of images is intended to obscure the identity of individuals and other identifying data, thus ensuring privacy protection. The process includes the development of a traffic recording system using two Raspberry Pi cameras, which simultaneously recorded traffic - one providing blurred and the other sharp images. The images were then aligned using the SIFT algorithm. Clear images were labeled using the DETR algorithm, with the labels being the same for blurred images as we recorded with two cameras. The final result is a database of over a thousand blurred images with labels in COCO format. This dataset is prepared for use in further research and development of traffic detection algorithms with built-in privacy. Detekcija prometa zasebnost zamegljene slike Raspberry Pi DETR format COCO baza slik Traffic detection privacy blurred images Raspberry Pi DETR COCO format image dataset true false false Slovenski jezik Angleški jezik Diplomsko delo/naloga 2023-07-10 12:55:00 2023-07-10 12:55:06 2023-07-11 11:07:12 0000-00-00 00:00:00 2023 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 59341 158433539 Kreft_Jakob_-_Vgrajena_zasebnost_pri_videonadzoru_prometa.pdf Kreft_Jakob_-_Vgrajena_zasebnost_pri_videonadzoru_prometa.pdf 1 851F4E32BA40789F353185AA503E296D 2ee31e50fb94904d7a29c22257ab92e8525fee2ba09bfc9618fb9d1e2e61e872 cdc6be96-1f0f-11ee-a314-00155dcfd717 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=171403 Fakulteta za elektrotehniko 0 0 0