A novel multimedia-forensic analysis tool (M-FAT)

Digital forensics has become a fundamental requirement for law enforcement due to the growing volume of cyber and computer-assisted crime. Whilst existing commercial tools have traditionally focused upon string-based analyses (e.g., regular expressions, keywords), less effort has been placed towards the development of multimedia-based analyses. Within the research community, more focus has been attributed to the analysis of multimedia content; they tend to focus upon highly specialised specific scenarios such as tattoo identification, number plate recognition, suspect face recognition and manual annotation of images. Given the ever-increasing volume of multimedia content, it is essential that a holistic Multimedia-Forensic Analysis Tool (M-FAT) is developed to extract, index, analyse the recovered images and provide an investigator with an environment with which to ask more abstract and cognitively challenging questions of the data. This paper proposes such a system, focusing upon a combination of object and facial recognition to provide a robust system. This system will enable investigators to perform a variety of forensic analyses that aid in reducing the time, effort and cognitive load being placed on the investigator to identify relevant evidence.