Here I will present a few of the project and research topics I have been working on. For most projects I provide an abstract that can be downloaded. If you would like to review the full report then contact me and I can send it!
Please note that all text is covered by copyright laws.
The projects that are covered are:
- Automatic determination of eye-fixation location
- Eye tracking
- ERBN modelling in games
- Knowledge retrieval
- Who's talking; automated speaker recognition
- KneeTwister, laxity and angle measurement
This abstract is taken from the thesis Automatic determination of eye-fixation location, date: February 02, 2009. The thesis is supervised by Dr.ir. J.W.H.M. Uiterwijk, Prof.dr. E.O. Postma, Dr. R.L. Westra, S.M.H. Jansen, M.Sc. The thesis submitted in partial fulfilment of the requirements for the degree of Master of Science of artificial intelligence at the faculty of humanities and sciences of Maastricht University.
Authors: Drs.ing. P. Hanckmann
This is the integral version of the report Eye tracking, written at January 21, 2008. It represents the results of of the research done for the course "Additional component computer graphics" at the Eindhoven University of Technology (TU/e) Faculty of Mathematics and Computer Sciences. It was supervised by Dr. ir. H.M.M. van de Wetering.
Authors: R. van Brakel, P. Hanckmann
This abstract is taken from the report ERBN modelling in games, written at June 21, 2007. It represents the results of of the research done in block 4.3 at the Maastricht University Faculty of Humanities and Sciences.
The report describes the results of the AI master project at the University of Maastricht. It covers the development of the language to describe Entity Relational Bayesian Networks (ERBN) and the tools to use this language.
Authors: P. Hanckmann, M. v.d. Heijden, P. Mertens, S. Uyttendaele
This report is written at June 1, 2007. It represents the results of a research project at the Maastricht University, Knowledge Engineering department.
The research was based on tools that can process documents (diaries) and place them in a document space. This high dimensional document space had to be reduced to a three dimensional space and projected on a sphere. With all the documents represented on a sphere it should be possible to see clusters of documents, zoom in on them and to see more detail.
Authors: M. de Mulder, C. Porteners, L. Ramekers, P. Hanckmann, P. Mertens
Who's talking; automated speaker recognition
This abstract is taken from the report Speaker recognition, Who's talking? written at January 24, 2007. It represents the results of a research project at the Maastricht University, Knowledge Engineering department.
The research was based on automated speaker recognition in single and multi speaker environments. All software was written in MatLab. And the research was based on different recognition systems (Gaussian Mixture Model, Neural Networks and Nearest Neighbours while using cc, mfcc and lpc as methods to extract the coefficients).
Authors: D. Bloembergen, ing. P. Hanckmann, A. Lautenbach, B. Mehlkop, F. Schadd
KneeTwister, laxity and angle measurement
This abstract is taken from the report KneeTwister, laxity & angle measurement written at June 1, 2005. The report is written under supervision of dr. ir. P. Holmes and ing. F. Cook. It represents the results of device developed and implemented at the Unit of joint reconstruction, RJAH Orthopaedic Hospital, Oswestry, United Kingdom.
The KneeTwister is a device which will be used to measure the laxity in the knee joint. It also measures the angle between the upper and lower leg during an internal and external rotation.
Authors: ing. P. Hanckmann , ing. E. Mulders