From Physics to Forensics


I first became interested in forensic science in 1987 when I was a physics student at the University of Applied Sciences in Rijswijk. I did an internship in the tool-mark and firearm department, and although I really liked the work there, when I graduated there were no vacancies at the laboratory. Working in an industrial environment appeared more attractive to me than working for the government at that time, so I started my working life at Oce Nederland B.V., conducting research on digital photocopying machines.

Some 3 years later, vacancies were advertised in the physics department of the Netherlands Forensic Institute (NFI). My desire to work in the forensics field was undimmed. Furthermore, the experience in image processing that I had gained at Oce was useful, because we were developing databases on shoe prints and tool marks at NFI. I decided to specialize in tool-mark and shoe-print investigation, also receiving training as an expert witness in this field. After 3 years, I gained Dutch court approval as an official expert witness.

A reorganisation of the institute in 1995 allowed me to use my image-processing knowledge as part of a new group in the institute’s firearm department. At that time, commercial systems were being developed for the automatic comparison of cartridge cases and bullets. In order to evaluate these systems, I conducted a research project on comparison algorithms. As a second research project, I started working with the pathology departments of NFI and the University of Amsterdam on the use of computerized tomography scans for determining the caliber of a bullet in a living person.

The first time I actually had to appear as a witness in court is still clear in my memory, partly because the so-called “ballpoint pen case” drew a lot of publicity. A woman in Leiden, the Netherlands, had been killed with a ballpoint pen, which penetrated her brain through her eye. The suspect told his psychiatrist that he had killed his mother with a crossbow and this ballpoint pen and was convicted. However, the case was reinvestigated when it reached the appeals court. The father, and several experts, did not believe that the woman could have been killed with a crossbow.

This is where the forensic experts came into play. We conducted ballistic experiments in gelatin with a high-speed video camera in order to examine the condition of the ballpoint pen after it had been shot with the crossbow. Under special conditions of the crossbow, it appeared that it was in fact possible to fire the ballpoint pen without damaging it, implying that the scenario was technically possible. Challenging experiments had to be done in a short time under high pressure.

During the ’90s, the field of image processing and pattern recognition in forensic science developed swiftly, and I had to testify several times in court on the investigation of video evidence. In 1999, the Netherlands Forensic Institute was formed by merging the Netherlands Forensic Laboratory and the Laboratory for Forensic Pathology, and I became a research and development coordinator focusing on developments in digital evidence at the new NFI. Later, this role was expanded to include R&D coordination for the Dutch High-Tech Crime Units of the police.

This move meant I was less involved in casework, so I had time to write my doctoral thesis, “Content-Based Information Retrieval from Forensic Image Databases,” and I successfully defended it in June 2002. In the future, I expect to work more in image processing and biometric devices. My academic colleagues think of my move to forensics as something that is really appropriate to me as a person.

The state(s) R&D is in

Perspective: Put Integrity High on Your ToDo List