ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Detecting traffic signs with a programme on the Android operating system

Miha Kramberger (2016) Detecting traffic signs with a programme on the Android operating system. EngD thesis.

[img]
Preview
PDF
Download (1608Kb)

    Abstract

    The objective of the diploma thesis is to develop a speed traffic signs detection application on the Android platform, which could be extended to other mobile operating systems with minimal adjustments. The programme is intended for all vehicles and other transport means in traffic, where the tablet or phone can be assembled so that does not disturb the driver at driving. The programme is appropriate for drivers who have not had similar systems installed in their cars. The diploma thesis presents a programme that is mostly prepared in the Java programming language for the Android operating system, used tools and a short description as well as procedures required for proper operations of the programme. The thesis also explains the sign detection issues and methods used to enable detection, as well as solutions used in the existing systems. Signs are detected by a camera set on the device; acquired data are processed by the OpenCV library. The programme does not use other technologies for detection and is not connected to the GPS system. The programme is intended for a long-term use in the vehicle, therefore, the battery autonomy is provided with the appropriate selection of the background colour that covers most of the screen, and with imaging in lower resolution. During the use, the driver set the device in a holder and turns it in the direction of driving so that the camera is not covered and slightly directed to the right side facing the signs.

    Item Type: Thesis (EngD thesis)
    Keywords: Android programme, sign detection
    Number of Pages: 34
    Language of Content: Slovenian
    Mentor / Comentors:
    Name and SurnameIDFunction
    doc. dr. Tomaž Dobravec258Mentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=51012&select=(ID=1536825027)
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 3259
    Date Deposited: 29 Feb 2016 15:48
    Last Modified: 17 Mar 2016 12:54
    URI: http://eprints.fri.uni-lj.si/id/eprint/3259

    Actions (login required)

    View Item