20.500.12556/RUL-119368 Sistem za meritev hitrosti kolesarjev s pomočjo analize videoposnetka System for measuring cyclist travel speed from a video Kolesarska infrastruktura je v nekaterih slovenskih mestih zelo slabo razvita. Za ustrezne izboljšave in ugotavljanje problematičnih odsekov potrebujemo analize trenutnega stanja. Eden od pomembnih podatkov pri tem je povprečna hitrost kolesarjev. Naprave za merjenje hitrosti, vgrajene v infrastrukturo, so pogosto drage in jih ni mogoče hitro in enostavno namestiti na zanimiva mesta. Eno od alternativ ponuja analiza slike oziroma videoposnetka. Namen naloge je zasnovati in izdelati programsko opremo za ugotavljanje povprečne hitrosti kolesarjev z videoposnetka na krajših odsekih cest ali kolesarskih stez. S tem bomo dosegli cenejše in hitre analize brez poseganja v trenutno kolesarsko infrastrukturo. Cilj naloge je izdelana programska oprema, ki jo bo mogoče uporabiti na videoposnetkih prostega prometnega toka, posnetih sproti ali predhodno, in z njo ustrezno natančno določiti hitrosti posameznih kolesarjev na posnetku. The cycling infrastructure in some Slovenian cities is very poorly developed. For correct improvements and identification of problematic sections, we need to analyse current situation. One of the most important data is the average speed of cyclists. Speed measuring devices integrated in the infrastructure are often expensive and cannot be quickly moved or installed in interesting places. One of the alternatives is image or video analysis. The purpose of the task is to design and manufacture software to determine the average speed of cyclists from a video on shorter sections of roads or cycle paths. This will achieve cheaper and faster analyses without interfering with the current cycling infrastructure. The goal is to create software that can be used on traffic videos recorded in real time or in advance, and with them accurately determine the speeds of individual riders in the video. sistem za meritev hitrost kolesarjev analize videoposnetka umetna inteligenca OpenCV analize slike Python system for measuring cyclist travel speed video analysis artificial intelligence OpenCV image analysis Python true false false [K. Gomilšek] Slovenski jezik Angleški jezik Diplomsko delo/naloga 2020-09-08 08:33:38 2020-09-08 08:33:40 2022-08-31 03:52:46 0000-00-00 00:00:00 2020 0 Ljubljana 0 XXII, 57 str. 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 629.322:531.767:004.8(043.2) 223739 30525443 5854.pdf 5854.pdf 1 687E49968D8D7931158FB92EE51BDAE0 ce9bebc262b79210b2e85d240ad953c885d01bf4d3be3caf9bca7e5bf9487bb3 cf18ef3d-a1b9-11eb-a523-00155dcfd717 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=134074 Fakulteta za strojništvo 0 0 0