Dissertation/Thesis Abstract

Digital Image Source Authentication
by Schippers, David A., D.Sc., Capitol Technology University, 2018, 223; 27828526
Abstract (Summary)

The purpose of this quantitative correlational study was to determine if a unique signature exists for EXIF JPEG elements from an iPhone 7 iOS 11 for source device and native picture application authentication. Law enforcement investigators need reliable and field applicable techniques for source authentication (Kraetzer & Dittmann, 2015), facilitating criminal identification (Miller, 2013). The general problem was law enforcement does not possess the knowledge or expertise for digital image source authentication on specific hardware and version implementations (Cummins Flory, 2016; Gogolin, 2010; Holt & Bossler, 2016). The specific problem was law enforcement does not possess the knowledge or expertise for digital image source authentication with EXIF JPG images for an iPhone 7 with iOS 11 (Cummins Flory, 2016). Literature review presented law enforcement training considerations, hardware/software implications for EXIF JPEG source authentication, and non-EXIF source authentication implications. This quantitative correlational study presented a moderate to strong correlation between Manufacturer, Model and Operations to a unique signature from EXIF JPEG. Test-retest reliability identified secondary quantization table usage, reducing anticipated 100% matching output to 99.04%. Field values for an iPhone 7 with iOS 11 were identified with differences from iPhone 5s, iPhone 6 and Samsung Galaxy S7 with Android 7.0.

Indexing (document details)
Advisor: Beatty, Dennis
Commitee: Butler, William, Pittman, Jason
School: Capitol Technology University
Department: Technology (PhD)
School Location: United States -- Maryland
Source: DAI-B 81/9(E), Dissertation Abstracts International
Source Type: DISSERTATION
Subjects: Information Technology, Computer science, Computer Engineering
Keywords: EXIF, Image forensics, JPEG, Metadata, Quantization, Source authentication
Publication Number: 27828526
ISBN: 9781658487986
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest