Samodejno odpiranje garažnih vrat s prepoznavo registrskih tablic

OMERZU, MATEJ (author), Burnik, Urban (mentor)

V diplomskem delu postavimo sistem za video nadzor, ki s pomočjo sistema za prepoznavo registrskih tablic avtomatsko odpira garažna vrata. Sestavljajo ga platforma Home Assistant, sistem za video nadzor Frigate in programska oprema za prepoznavo registrskih tablic CodeProject.AI. V uvodnem delu naloge si ogledamo in primerjamo najbolj priljubljene rešitve na trgu za sisteme pametnega doma, video nadzora in prepoznave registrskih tablic. Predstavimo še delovanje protokola MQTT, ki ga bomo uporabili za izmenjavo podatkov med temi tremi sistemi. V nadaljevanju opišemo uporabljeno strojno opremo in po korakih predstavimo postavitev programske opreme sistema. Najprej opišemo namestitev platforme za pametni dom Home Assistant, v katerem namestimo še sistem za video nadzor Frigate in programsko opremo za krmilnik garažnih vrat. Sledi še namestitev programske opreme za prepoznavo registrskih tablic CodeProject.AI in Docker kontejnerja, na katerem teče skripta, ki povezuje sistema za video nadzor in prepoznavo registrskih tablic. Nato ustvarimo še scenarij avtomatizacije. Na koncu še preverimo in predstavimo delovanje sistema. Poleg tega navedemo tudi glavne težave pri delovanju in njihove rešitve. Zaključimo z ovrednotenjem sistema in pregledom možnih izboljšav za prihodnost.

• Language: Slovenian
• Keywords: Home Assistant, Frigate, CodeProject.AI, MQTT, hišna avtomatizacija, prepoznava registrskih tablic
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FE - Faculty of Electrical Engineering
• Year: 2024
• PID: 20.500.12556/RUL-165618
• COBISS.SI-ID: 221042947
• Publication date in RUL: 10.12.2024

Secondary language

• Language: English
• Title: Automatic Garage Door Opening Using License Plate Recognition

Abstract

In this thesis, we set up a video surveillance system that automatically opens garage doors using a license plate recognition system. The system consists of the Home Assistant platform, the Frigate video surveillance system, and the CodeProject.AI license plate recognition system. In the introductory part of the thesis, we compare the most popular systems for smart home, surveillance, and license plate recognition. We also present the operation of the MQTT protocol, which we use for the data exchange between these systems. Next, we describe the used hardware and present the step-by-step set-up of the system software. First, we describe the installation of the Home Assistant smart home platform, in which we also install the Frigate video surveillance system and software for the garage door opener. This is followed by the installation of the software for license plate recognition CodeProject.AI and a Docker container, which runs the script that connects the surveillance system with the plate recognition software. We then create the automation scenario. Finally, we assess and present the operation of the system. Additionally, we list the main problems we encounter during operation and their solutions. We conclude with an evaluation of the system and an overview of possible improvements for the future.

• Keywords: Home Assistant, Frigate, CodeProject.AI, MQTT, home automation, license plate recognition