Automatic Full and Partial Shoeprint Retrieval System for Use in Forensic Investigations

Forensic identification methods can be divided into physical forensics, chemical forensics, and biological forensics categories. Robust automatic identification or retrieval systems, such as AFIS and DNA-STR analysis systems, for fingerprints and biological evidence have been introduced. However, there are still challenges for shoeprint evidence. For example, most of identification or retrieval methods can only be applied to intact shoeprints, and do not compare shoeprints with different sizes. In addition, the shoeprints at a real crime scene are commonly incomplete, noisy, and unclear. In this paper, we propose a new shoeprint retrieval system that normalizes the size of shoeprints and is efficient in retrieving full- and partial-print shoeprints. Experimental results demonstrate our proposed method is capable of handling shoeprint identification in varied situations such as dusting environment, partial print, etc. In particular, our proposed method outperforms Ho's method when dealing with different shoe sizes.