20.500.12556/DKUM-96004 Pametni video nadzor domačega okolja z uporabo nevronskih mrež magistrsko delo Smart video surveillance of the home environment using neural networks V magistrskem delu smo razvili pameten sistem video nadzora domačega okolja, ki s pomočjo nevronskih mrež omogoča zaznavanje oseb in vozil v realnem času. Sistem uporablja model YOLO za prepoznavanje objektov, lasten model na osnovi ResNet-50 za prepoznavo znamke in modela vozila ter OCR za branje registrskih tablic. Zajeti podatki se shranjujejo v MSSQL bazo in so dostopni prek spletnega vmesnika, razvitega v .NET in Angularju. Rešitev omogoča pregled dogodkov, arhiviranje posnetkov in prikaz v živo ter izkazuje zanesljivo delovanje v različnih vremenskih in svetlobnih pogojih. In this master’s thesis, we developed a smart home video surveillance system that uses neural networks for real-time detection of people and vehicles. The system employs the YOLO model for object detection, a custom ResNet-50–based model for recognizing vehicle brands and models, and OCR for license plate reading. Captured data are stored in an MSSQL database and accessed through a web interface built with .NET and Angular. The solution enables live view, event tracking, and video archiving, demonstrating reliable performance under various weather and lighting conditions. Pametni video nadzor nevronske mreže prepoznavanje vozil YOLO Smart video surveillance neural networks vehicle recognition YOLO true true false [U. Kos] Slovenski jezik Angleški jezik Magistrsko delo/naloga 2025-11-20 10:18:57 2025-12-22 10:00:45 2026-01-31 03:15:38 0000-00-00 00:00:00 2025 0 Maribor 0 Maribor 1 spletni vir (1 datoteka PDF (XI, 63 str.)) 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.032.26:004.932(043.2) 266779651 MAG_Kos_Uros_2025.pdf MAG_Kos_Uros_2025.pdf 1 BA46D001DEEE93FD264DB4692FBC01AC eeb9c46fabb45d3ecc4a146509fea3073515391ae2f29f35db66b4910dd31cc9 c36ab636-c5f6-11f0-815e-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=200784 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0