Home

Welcome on the webpage of Rob Heylen. I am a researcher working in the field of hyperspectral image processing since 2009. I have worked at the University of Antwerp, Belgium, under Prof. Paul Scheunders, and at the University of Florida under Prof. Paul Gader. Currently, I am holding a FWO postdoctoral position, and am associated with the University of Antwerp.

My main field of interest is unmixing of hyperspectral images, although I have worked in several other areas, such as statistical mechanics, computational physics, artificial intelligence and robotics. In recent years, I have worked on
  • Estimation of the number of endmembers in hyperspectral images
  • Endmember extraction under linear or nonlinear mixing assumptions
  • Several algorithms for solving the inversion or unmixing step, which requires solving a constrained least-squares problem
  • Development of a distance-geometric interpretation of the unmixing chain, allowing flexible use of different metrics or preprocessing steps
  • Several applications of unmixing, such as water quality retrieval and detection of the adjacency effect in hyperspectral images.
Furthermore, I have written a review paper on nonlinear hyperspectral unmixing techniques, which I presented in the form of a half-day tutorial in IGARSS 2013.

On this webpage, you can find most of my papers, and the code of several of the algorithms we have developed over the years. If you do not find what you are looking for, or have questions concerning some of the techniques and algorithms, you can contact me at rob.heylen@uantwerpen.be