Application of line scan devices in optical flow based image processing

OData support
Dr. Vajta László
Department of Control Engineering and Information Technology

For my master’s thesis, my task was to examine the feasibility of a sensor investigating the road surface by means of a linear imaging system. This sensor would be capable of doing so in the direction of the moving car onto which it is mounted. For this purpose an extensive literature research has been carried out in the field of methods used for the investigation of surfaces. The findings of this research are presented in this paper with detailed description of the optical methods, especially the ones that make use of linear images and their processing possibilities.

If successful, the examined device may be a part of a crowdsourcing-based system, in which the moving data-collector agents (vehicles) transmit the captured image flows and data about the road surface to a cloud, where these are processed and saved into a road databank. The results of the research carried out on the currently available road quality investigation systems as well as the existing databanks are presented in this work.

The findings of my research in the field of driver assistance systems presently available on modern cars - whose existence is crucial for the crowdsourcing aspect of the aforementioned system - are also detailed in this paper. Furthermore, I introduce the VehicleICT platform, the essential architecture for creating the crowdsourcing project.

In my thesis I explore the limitations of a road surface investigating device and present the simulation prepared for the examination of its qualities. Additionally, I demonstrate the design procedure and operation of a functional model of such equipment. The designed device is then compared to those systems that are commercially available today.


Please sign in to download the files of this thesis.