Date of Award

10-1-2021

Publication Type

Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

First Advisor

M.A.S Khalid

Second Advisor

R. Ruparathna

Third Advisor

M. Azzouz

Keywords

Deep learning, Light detection, Traffic sign

Rights

info:eu-repo/semantics/openAccess

Abstract

Traffic sign and light detection are core components of Advanced Driver Assistance Systems (ADAS) and self-driving vehicles. The automotive industry is widely employing numerous approaches for automation through computer vision techniques. Object detection algorithms based on deep learning can be divided into two main categories, two stage and single stage detection algorithms. Two stage algorithms are designed to improve detection accuracy. While single stage algorithms are constructed to be faster, this increases their suitability for real time applications. This thesis presents a lightweight traffic sign and light detector by adapting a single stage, Single Shot Multibox Detection (SSD) algorithm by providing both high accuracy and real time detection capability. Therefore, the Visual geometry group (VGG16) base network in original SSD is replaced by MobileNet, that expertly manages detection speed and network size because of its lighter architecture. It is essential for the application domain to be able to detect small objects which is what the original SSD struggles with. For autonomous driving the detection results with respect to the distance of an object is of particular interest. A comfortable braking distance is needed in case of traffic signs and lights. That requires object detection from a farther distance, but farther distance makes the object to be detected appear smaller. Thus, this work further optimizes the number of feature map layers of the algorithm for the detection of small objects along with a better trade off between accuracy and detection time. Experimental results confirm the effectiveness of the proposed model as compared to the standard SSD with VGG16 and SSD with MobileNet V2.

Share

COinS