| |Home |About |Interests |Books |Links |Fun |Photos |Software |Contact | |
From 1996 to 1999 I worked with remote sensing and image
processing at former Satellitbild (now a branch of the
Swedish National Land Survey
in the winter town of Kiruna. In 1999 I moved to
Stockholm and spent two years at
Carlstedt Research & Technology as a consultant.
In 2001 I started to work with face tracking
at SmartEye AB, a spin-off company from CR&T. Working with
real-time tracking and pattern recognition
was a challenge and great fun, but all good things come to an end and I spent 2005-2014 working with tracking, sensor modelling and various
R&D projectss for manned and unmanned aerial aircraft for
Saab Aeronautics. Here is a detailed CV.
From 2014 to 2016 I did some consulting for the Alten group,
and as of June 1 2016 I have moved to Germany (Langen, outside Frankfurt) and started working for
Panasonic working with self-parking cars and ADAS.
Short background:
B.Sc. in physics from Uppsala University, 1986 (theoretical physics and astronomy).
6 months as an exchange student at Sussex University, spring 1985. Project work in Kaluza Klein Theory - manual tensor crunching that Mathematica or Maple would do better and with less mistakes - and an essay on black holes. For what it's worth, I've made a rough digital version of the kaluza-klein document here, typed on semi electronic typewriter 1985, complete with spelling (and possibly conceptual) mistakes.
Programmer at the department of Clinical Neurophysiology, Akademiska Sjukhuset, Uppsala. Various programming tasks and numerical analysis (1985-1986).
Ph.D. student at the Onsala Space Observatory (1986-1988).
Ph.D. thesis (1994) at the department of Signals and Systems (formerly called Applied Electronics) , Chalmers University of Technology. Teaching in analog/digital signal processing, doctorate work on a system for automatic delineation of large scale objects in medical images, involving computer vision, signal processing, machine learning and AI methods. The thesis topic was a system for automatic segmentation, applied to ultrasound and gamma camera images of the heart. The image was represented as a resolution pyramid with feature vectors at every resolution element (pixel), the most useful image primitive being "blobs", or inflated local minima/maxima. The blobs were labeled with a relaxation labeling procedure, and pruned according to size, context and proximity to promiment edges. A top-level scheduler (really a finite state machine) was used in a processing and evaluation cycle to yield the most consistent scene interpretation. The blobs were "inflated" using a gray-scale based medial axis transform to yield the final result on a coarse resolution level. All models - scene and process model - were exchangeable and three different types (CAT and gamma camera images) of medical imagery were tested. Parts of the thesis available here (rar-archive with pdf-files).
Post doc position at KFA, Jülich, Germany with Frank Pasemann and his group (ATÖ). I Spent most of the time learning about neurobiology of vision.
Stipends from Hellmuth Hertz Foundation and the Technical Research Council (now Vetenskapsrådet) for one year's stay at Inria, Sophia Antipolis, north of Antibes on the Riviera. I Learned about Unix and C++ and became a "tourist" in areas such as artificial life, blackboard systems, watershedding algorithms at the PASTIS project, as of 1997 replaced by the ARIANA project.
Selected References:
Molander S. & Broman H., Gustavsson T. (1991): "A controllable Medial Axis Transform", Proceedings of the society for automated image analysis (now SSBA), Linköping, 6-7 March, pp. 240-243.
Molander S. & Broman H. (1993):"Knowledge-based segmentation and state-based control in image analysis: two examples from the biomedical domain", Signal Processing (Elsevier), special issue on intelligent systems for signal and image understanding, Vol. 32, Nos 1-2., May, pp. 201-215.
Molander S. (1995):"Blob analysis of biomedical image sequences: a model-based and an inductive approach", proceedings of the summer school in Huddinge (ed. Stig Andersson) Analysis of Dynamical and Cognitive Systems, Aug. 9-14 1993, in Lecture Notes in Computer Science, Springer verlag, Vol. 888,, pp. 169-187.
Molander S. (1994): "What is lacking in today's vision systems?", Proceedings of the Swedish Society for Automated Image Analysis (SSAB), Halmstads, March, 8-9, pp. 13-17.
Molander S. (2023): Dense Random Texure Detection using Beta Distribution Statistics (arxiv.org)
L. Golla, S. Molander, and D.-V. Nguyen. (2023): " Moving object detection with photometric monocular slam on a moving ego-platform. IEEE 26th International Conference on Intelligent Transportation Systems, doi=10.1109/ITSC.2015.267, 2023.
| |Home |About |Interests |Books |Links |Fun |Photos |Software |Contact | |
