Comparing Footwear Impressions that are Close Non-Matches Using Correlation-Based Approaches

Published

May 1, 2021

Author(s)

Gautham Venkatasubramanian, Vighnesh Hegde, Sarala Padi, Hariharan K. Iyer, Martin Herman

Abstract

Forensic activities related to footwear evidence may be broadly classified into the following two categories: (1) intelligence gathering and (2) evidential value assessment. Intelligence gathering includes identifying the make and model of a shoe impression by comparing its design elements with a database of outsole designs of known make and model, linking footwear impressions from one crime scene to those from a different crime scene, linking suspects to crime scenes, and other activities that provide leads for investigators. Assessment of evidential value, as practiced in the United States, involves a trained footwear examiner evaluating the degree of similarity between a known shoe of interest (together with its test impressions) and footwear impressions obtained from a crime scene, by performing side-by-side visual comparisons. However, the need for developing quantitative approaches for expressing similarities during such comparisons is being increasingly recognized by the forensic science community. In this paper, we explore the ability of similarity metrics to discriminate between impressions made by a shoe of interest and impressions made by close non-matching shoes. Close non-matching shoes largely share the same design and size. Therefore the ability to effectively discriminate between them requires considering, either explicitly or implicitly, not only design and size, but also wear patterns and, to some extent, individual characteristics. This type of discrimination is necessary for assessment of evidential value. The similarity metrics examined in this paper are correlation-based metrics, including Normalized Cross Correlation, Phase-Only Correlation, AvNCC and AvPOC. The latter two metrics are based on features obtained from a convolutional neural network. Experiments are performed using Everspry impressions, FBI boot data that consist of impressions of boots worn by FBI trainees, and the West Virginia University footwear impression collection. The results show that Phase-Only Correlation performs as well as or better than the other metrics in all cases for the data sets we considered.

Citation

Journal of Forensic Sciences

Volume

66

Issue

3

Pub Type

Journals

Download Paper

https://doi.org/10.1111/1556-4029.14658

Local Download

Keywords

footwear impressions, footwear evidence, shoeprints, similarity metrics, correlation matching, deep learning, convolutional neural network

Citation

Venkatasubramanian, G. , Hegde, V. , Padi, S. , Iyer, H. and Herman, M. (2021), Comparing Footwear Impressions that are Close Non-Matches Using Correlation-Based Approaches, Journal of Forensic Sciences, [online], https://doi.org/10.1111/1556-4029.14658, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=930527 (Accessed December 21, 2024)