Home // EMERGING 2019, The Eleventh International Conference on Emerging Networks and Systems Intelligence // View article

Semi-Automated Footwear Print Retrieval Using Hierarchical Features

Authors:
Tim vor der Brück
Thomas Stadelmann

Keywords: Footwear print; retrieval; tree edit distance

Abstract:
Footwear prints are one of the most commonly found pieces of evidence on crime scenes. They can be used both to connect different crimes with each other and to give important clues to the identity of the culprit. In many cases, these footwear prints are distorted or incomplete, which makes fully automated approaches for their identification and comparison unreliable. Hence, we propose a semi-automated approach, where a representation of key features is obtained manually by forensic experts, the comparison with outsole models from a database is done by a computer. To account for potentially bad quality footwear print pictures, we introduce a hierarchical fuzzy search that ranks the outsole models according to their degree of correspondence with the features of the footwear print. Furthermore, we conducted an evaluation that demonstrated the usefulness of the proposed approach.

Pages: 16 to 22

Copyright: Copyright (c) IARIA, 2019

Publication date: September 22, 2019

Published in: conference

ISSN: 2326-9383

ISBN: 978-1-61208-740-5

Location: Porto, Portugal

Dates: from September 22, 2019 to September 26, 2019