Detection of second-generation images using an assessment criteria method

Abstract

Purpose: The ability to distinguish between “original” and “copied” images has been a persistent forensic imaging difficulty and can be of some importance to certain criminal and civil investigations. The purpose of this paper is to introduce a novel assessment criteria method that incorporates visual and metadata-based information for the purpose of determining whether images are original or second-generation duplicates (copies made by rephotographing the original hardcopy). Design/methodology/approach: The study reflects difficulties raised from forensic cases and is modelled on fraud investigation that involved images sourced from camera phones. The method involved a new assessment-based criteria approach and the results were evaluated through their application to a sample set of second-generation images. Findings: The evaluation confirmed the validity of several theorised detection artefacts resulting in the articulation and presentation of 17 detection criteria considered useful for supporting image analysis. Originality/value: The result of this study is an expansion of the tools available to examiners for addressing complex image authentication problems. The criteria approach also assists with transparently communicating the details of the photo interpretation processes for review and scrutiny.