Traffic Sign Detection using Deep Neural Netwokrs

OData support
Supervisor:
Szemenyei Márton
Department of Control Engineering and Information Technology

Nowadays, autonomous vehicles are in the spotlight. Numerous IT companies and automobile manufacturers are working with great efforts to revolutionize the 21st century’s personal transportation. Computer vision and machine learning are important part of this revolution, partially because of their image and video processing and analyzing capabilities.

Partly because of the increasing quantity of data, deep learning has become one of the most important and fastest growing research areas in machine learning. Deep learning allows computational models that are composed of multiple processing layers to learn complex representations of data. Deep convolutional networks have brought about breakthroughs in image and video processing.

In this thesis, I present methods for creating training and test dataset for both classification and object detection problem. Furthermore, I show that convolutional networks can classify images with great accuracy. Moreover, the thesis demonstrates the capabilities of real-time object detection with deep learning models.

Our results show that deep learning is suitable for solving computer vision problems. It can robustly classify and detect objects in images.

Downloads

Please sign in to download the files of this thesis.