A translational and rotational invariant descriptor for automatic footwear retrieval of real cases shoe marks

Shoe marks found on the crime scene are invaluable for the identification of the culprit when no other piece of evidence is available. Thus semi-automatic and automatic systems have been recently proposed to find the make and model of the footwear that left the shoe marks. The systems proposed up to now have two main drawbacks, as they (i) are generally not based on rotation and translation invariant descriptions, and (ii) are tested on synthetic shoe marks, i.e. on shoeprints with added synthetic noise. Here we show the results of a translation and rotation invariant description based on the Fourier transform properties: the test is made on both synthetic and real shoe marks and a comparison with algorithms proposed by others is presented.