Research
Video-surveillance systems are one of the main source of information during investigations, thanks to their wide-spread and increasing presence in our countries, although the adopted closed-circuit devices are often affected by poor quality. Nonetheless, low quality images can give useful information both during the first phase of the investigation and in courtrooms
Also the digital documentation made during the crime scene analysis can be analyzed and processed in order to obtain valuable information for the investigation, e.g. thanks to pattern recognition techniques. Below you find our research aimed at the retrieval of the footwear employed during a crime.
Forensic Image Processing
Shoeprints found on the crime scene are useful to understand the crime dynamics, currently some systems have been proposed to automatically identify the make and model of the shoe that left the mark on the crime scene.
However these systems have been tested on shoe marks synthesized artificially
with different noise adding techniques. Here we present an image retrieval algorithm which combines the information of the phase of the Fourier transform of the shoe mark images with the power spectral density of the Fourier transform calculated on their Mahalanobis map, Fig. 1.
Differently from others, the algorithm performance is tested on real shoe marks coming from crime scenes. The proposed method is compared with other works, Fig. 2, and some preprocessing operators are also introduced and selected to reduce noise and enhance the matching probability, Fig. 3.
Fig. 2
Fig. 3