Prepoznavanje prometnih znakov z uporabo globokih konvolucijskih nevronskih mrež

Karamatić, Boris (author), Skočaj, Danijel (mentor)

Problema detekcije in prepoznavanja prometnih znakov postajata pomemben problem pri razvoju samovozečih vozil ter naprednih sistemov za asistenco vozniku. V diplomski nalogi bomo razvili sistem za detekcijo in prepoznavanje prometnih znakov. Za problem detekcije bomo uporabili značilnice iz agregiranih kanalov, za problem prepoznavanja pa globoko konvolucijsko nevronsko mrežo. Opisali bomo kako konvolucijske nevronske mreže delujejo, kako so zgrajene ter razložili pomen posameznih nivojev. Razložili bomo pristop, katerega smo uporabili pri razvoju konvolucijske nevronske mreže. Končne rezultate detekcije in klasifikacije bomo evalvirali na uveljavljenih bazah prometnih znakov.

• Language: Slovenian
• Keywords: konvolucijska nevronska mreža, prometni znaki, detekcija, prepoznavanje, klasifikacija
• Work type: Bachelor thesis/paper
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2016
• PID: 20.500.12556/RUL-85567
• Publication date in RUL: 16.09.2016

KARAMATIĆ, Boris, 2016, Prepoznavanje prometnih znakov z uporabo globokih konvolucijskih nevronskih mrež [online]. Bachelor’s thesis. [Accessed 31 March 2025]. Retrieved from: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=eng&id=85567

Secondary language

• Language: English
• Title: Traffic sign recognition with deep convolutional neural networks

Abstract

The problem of detection and recognition of traffic signs is becoming an important problem when it comes to the development of self driving cars and advanced driver assistance systems. In this thesis we will develop a system for detection and recognition of traffic signs. For the problem of detection we will use aggregate channel features and for the problem of recognition we will use a deep convolutional neural network. We will describe how convolutional neural networks work, how they are constructed and we will explain the use of every layer. We will describe the steps we took to develop our convolutional neural network. We will evaluate the results of detection and classification on established traffic sign datasets.

• Keywords: convolutional neural network, traffic signs, detection, recognition, classification