Date of Award

11-23-2019

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

First Advisor

Kemal Tepe

Keywords

ADAS, License Plate Localization, Machine Learning, Monocular Vision, Neural Network, Vehicle Distance Detection

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

With the development of new cutting-edge technology, autonomous vehicles (AVs) have become the main topic in the majority of the automotive industries. For an AV to be safely used on the public roads it needs to be able to perceive its surrounding environment and calculate decisions within real-time. A perfect AV still does not exist for the majority of public use, but advanced driver assistance systems (ADAS) have been already integrated into everyday vehicles. It is predicted that these systems will evolve to work together to become a fully AV of the future. This thesis’ main focus is the combination of ADAS with artificial intelligence (AI) models. Since neural networks (NNs) could be unpredictable at many occasions, the main aspect of this thesis is the research of which neural network architecture will be most accurate in perceiving distance between vehicles. Hence, the study of integration of ADAS with AI, and studying whether AI can safely be used as a central processor for AV needs resolution. The created ADAS in this thesis mainly focuses on using monocular vision and machine training. A dataset of 200,000 images was used to train a neural network (NN) model, which accurately detect whether an image is a license plate or not by 96.75% accuracy. A sliding window reads whether a sub-section of an image is a license plate; the process achieved if it is, and the algorithm stores that sub-section image. The sub-images are run through a heatmap threshold to help minimize false detections. Upon detecting the license plate, the final algorithm determines the distance of the vehicle of the license plate detected. It then calculates the distance and outputs the data to the user. This process achieves results with up to a 1-meter distance accuracy. This ADAS has been aimed to be useable by the public, and easily integrated into future AV systems.

Share

COinS