Pose Estimation using Fiducial Markers

OData support
Kovács Viktor
Department of Automation and Applied Informatics

Image processing is a rather popular technology nowadays. Its methods are widely used from picture editing at home to several scientific fields, including healthcare (enhancing and reducing noise from medical images), security technology (biometric verification, for example fingerprint reading, face recognition, detecting intruders using motion detection), space science (reducing noise from the photos of space telescopes), robotics (localization) and in augmented reality software.

In my thesis I introduce different methods and algorithms for processing digital images, including object and shape recognition and separation, thinning lines, noise reduction and methods for creating binary images from coloured ones. I also introduce connected-component labelling algorithms on binary images and different popular methods for marker recognition.

After that, relying on the presented knowledge, I design and evaluate a program that can recognise quadrilateral markers on images that are opened on one side.


Please sign in to download the files of this thesis.