An automatic footwear retrieval system for shoe marks from real crime scenes

Shoeprints found on the crime scene are useful to understand the crime dynamics. Several semi-automatic and automatic systems have already been proposed to identify the make and model of the shoe that left the mark on the crime scene, however these systems have been tested on simulated shoe marks, i.e. not coming from real crimes scene but artificially synthesized with different noise adding techniques. In this paper, we propose descriptors based on appropriate information of shoeprint pattern in the frequency domain, to retrieve the footwear that produced the shoe mark. Suitable operators, as for example the Mahalanobis distance map, are then selected and involved in the calculation of these descriptors, in order to reduce noise and enhance shoeprint textures found in real cases. In fact and differently from others, the performance test of this algorithm is done on shoe marks coming from real crime scenes, and the results are promising.