Search Menu
Photonics Media Photonics Buyers' Guide Photonics Spectra BioPhotonics EuroPhotonics Vision Spectra Photonics Showcase Photonics ProdSpec Photonics Handbook
More News

Sole Searching

Facebook Twitter LinkedIn Email Comments
Sally B. Patterson

Scotland Yard may someday have a better way of tracking criminals — by examining the treads on their shoes. Recently, laws concerning police powers in the UK have changed to acknowledge footwear matching as a valid form of courtroom evidence comparable to fingerprints and DNA. This means that it is increasingly common for shoe imprints to be made when a suspect is taken into custody. However, human interpretation and cross-referencing of shoeprint data has been time-consuming, and computer matching resources, limited.


A system being researched for shoe-print analysis uses some 500 features to match patterns against established databases of sole types. Images courtesy of the University of Sheffield.

Now, researchers led by Nigel Allinson and Maria Pavlou at the University of Sheffield are working with support from the UK Engineering and Physical Sciences Research Council to develop a definitive automatic classification system that can rapidly identify sole markings and search databases for matching types. The methodology that they are developing analyzes the treads, searching for logos, ridges and shapes, and the size and layout of the patterns. It examines patterns at different brightness levels to determine maximally stable external regions, which are processed as key identifiers for use in comparing prints — even incomplete ones or those taken at different angles.

“We are now looking at finding the sufficient feature set for all shoes — currently looks like about 500 features,” Allinson explained. Using these to check features against existing commercial and governmental databases of footwear patterns, they achieve a better than 90 percent hit rate in locating the top five matches.

However, there is a lot of footwork ahead before a device can be commercialized for widespread use. “We are attempting to build up something as large and robust as the automatic fingerprint identification system, and there is no agreed set of features for footwear,” Allinson said. Pattern recognition techniques and processing software must be optimized.

The team also hopes to streamline the imprint technique, which consists of taking prints using vegetable dye or motor oil on sensitized paper and scanning them into the computer. It could be a couple of years before an instrument is available for use in custody facilities, Allinson said. “We then need to consider crime scene marks — a much more difficult task.”

Even so, the time will come when the technology could force would-be perps to do some soul-searching before putting their telltale soles up for scrutiny.

Photonics Spectra
Jun 2006
industrialLighter Sidemethodologyshoe-print analysisUniversity of Sheffield

back to top
Facebook Twitter Instagram LinkedIn YouTube RSS
©2019 Photonics Media, 100 West St., Pittsfield, MA, 01201 USA,

Photonics Media, Laurin Publishing
x Subscribe to Photonics Spectra magazine - FREE!
We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.