Bettina Selig

Stochastic Watershed

The stochastic watershed is a method recently presented that builds on the classical seeded watershed algorithm. It creates a probability density function for edges in the image by repeated applications of the seeded watershed with random seeds. We are investigating several ways of overcoming some obvious disadvantages of the method.

Image Analysis on Compression Wood

In cooperation with the Department of Forest Products at SLU we develop a method, which automatically segments and measures wood fibers (tracheids) in compression wood cross-sections.

The properties of wood fibers are important for the performance of wood and fiber products. Compression wood is of limited value in the forest product industry, since its fibers are stiffer and more difficult to collapse compared to normal wood. But because compression wood is difficult to be completely avoided, a better understanding of it is needed in order to develop mathematical prediction models of wood mechanics.

Journals (reviewed)

Bengtsson Bernander K, Gustavsson K, Selig B, Sintorn IM, Luengo Hendriks CL Improving the Stochastic Watershed Accepted by Pattern Recognition Letters.

Selig B, Luengo Hendriks CL, Bardage S, Daniel G, and Borgefors G Automatic measurement of compression wood cell attributes in fluorescence microscopy images In Journal of Microscopy, 2012, 246: 298-308.

Conferences (reviewed)

Selig B, Luengo Hendriks CL, Bardage S, and Borgefors G Segmentation of Highly Lignified Zones in Wood Fibre Cross-Sections In Proceedings of 16th Scandinavian Conference on Image Analysis (SCIA), Oslo, Norway, June 15-18 2009.

Others

Selig B, and Luengo Hendriks CL Stochastic watershed – an analysis In Proceedings of SSBA 2012, Symposium on Image Analysis, Stockholm, Sweden, March 8-9, 2012.

Selig B, Khan Niazi MK, and Nyström I Using a Ring-Shaped Region around the Optic Disk in Retinal Image Registration of Glaucoma Patients In Proceedings of SSBA 2010, Symposium on Image Analysis, Uppsala, Sweden, March 11-12, 2010.

Selig B, Luengo Hendriks CL, Bardage S, and Borgefors G Measuring Distribution of Lignin in Wood Fibre Cross-Section In Proceedings of SSBA 2009, Symposium on Image Analysis, Halmstad, Sweden, March 19-20, 2009.

Link to publications listed in the Uppsala University DiVA publication database.

Stochastic Watershed: A Comparison of Different Seeding Methods
by Karl Bengtsson Bernander and Kenneth Gustavsson

Fuzzy Segmentation of Wood Fiber Cross-Section Images
by Feng Sitao

Link to publications listed in the Uppsala University DiVA publication database.

Spring 2012: Image Analysis I
Lectures about segmentation and color.

Autumn 2011: Application Oriented Image Analysis (Post graduate course)
Lectures about segmentation and color.

Spring 2011: Image Analysis I
Lectures about segmentation and color.

Spring 2009: IT Project
Supervision of students building and programming a rescue robot with the aim to take part in the RoboCup German Open 2009. I was especially helping with the vision system.

Fall 2008: Computer Programming I
Lessons about programming using Java.