Oklahoma State Research Aids Forensic Investigators in Identifying Auto Paint

Oklahoma State University students are making a big difference in a national effort to identify criminals using a sample of automobile paint taken from a crime scene.

OSU chemistry professor Barry K. Lavine was awarded a grant to continue his research of automobile paint samples to benefit forensic scientists, the Stillwater News Press reported.

Ph.D student Collin White and undergraduate student Matthew Allen work with him in his chemistry lab on the fourth floor of the Physical Science building.

“These are local folks making a big difference,” Lavine said.

The grant has the potential to speed up the paint sample process using infrared spectral data. In order to do this, the data is obtained from analyzing automotive paint samples to narrowing down suspect vehicles quicker.

During a hit and run crime, for example, a small sample of paint from the suspect’s car is taken from the victim’s clothing. The sample is taken to a lab using analytical chemistry. Forensics uses a microscope to see the paint and its layers, then characterize it. A single sample can identify the make, model and year of the car.

Lavine said his goal is to identify 30 samples in two days.

White works with the computer side of things by comparing the spectral paint to the paint of the automobile using algorithms. White, a native of Waukomis, went to Enid Public Schools and graduated from OSU with an undergraduate degree in chemistry.

He is able to determine if the car comes from Oklahoma City, for example.

“We can tell the location by the different variations of paint,” he said.

All it takes is a small smear of clear-coat paint.

White said it’s the most informative layer. With the sample, algorithms can determine the car’s manufacturing plant group. There are about three to five manufacturing groups, and within those are up to eight plants.

He must determine the similarity of the average spectrum sample to the various groups.

Allen, a computer science major, compared the process to putting a puzzle together.

“When you start putting a puzzle together, you start with the edges because it has flat sides,” he said.

Allen is a programmer for the research, monitoring algorithms.

“I’m finding those flat sides, except it’s not something the human eye can see,” he said. “That is where the computer comes in.”

Allen worked on moving 12-year-old software to a new machine earlier this month.

The goal for the research is to develop it into something that is used throughout the field of forensic science. The $443,000 grant from the National Institute of Justice is Lavine’s third grant in four years. Since 2010, his research has collected more than $1.2 million in funding.