Digital evidence, also known as electronic evidence, is a rapidly expanding area of forensic science encompassing a diverse range of forensic disciplines. These include computer forensics, audio video analysis, and speaker and face comparison.
Computer forensics involves the extraction, preservation and analysis of data associated with computers, mobile phones and networks for a wide range of crimes from cyberbullying to fraud and international terrorism. Cybercrime is an area of increasing challenge for law enforcement. A wide range of forensic tools and techniques have been developed to recover, extract, preserve and analyse data. Examples of examinations include the recovery of deliberately erased data, the detection and tracking of unauthorized network intrusion and the identification of the origin (or source) of data through analysis of metadata and file artefacts.
Audio video analysis includes the recovery and extraction of evidence from recording devices, such as digital audio recorders and CCTV systems. It also includes subsequent conversion, enhancement and analysis of audio video data. CCTV is a powerful tool that assists police in the investigation of crimes ranging from theft to serious assault. Other examples of audio video analysis include speed estimation of vehicles from CCTV footage, authenticity analysis of audio recordings and comparison of a suspect’s clothes with that of an offender captured on CCTV.
Forensic speaker comparison generally involves the identification of an unknown person through analysis of voice characteristics. A combination of phonetic and computer based analyses are applied to the comparison of recorded speech utterances. Computer systems use many different pattern matching algorithms to compare speech. These systems may be text-dependent or text-independent. Typically, unknown speech is profiled through spectral analysis and modelling. The unknown profile is then used to search a database of voices resulting in a list of candidates who share speech characteristics. Computer speech recognition systems are most reliable when applied to verification of a speaker in a security context. Speech utterances can also be compared through phonetic and linguistic analysis. In this case, speech sounds and phrasing are profiled and compared by an expert through critically listening often in conjunction with spectral analysis.
Face comparison involves the identification of an unknown person through comparison and analysis of facial images. A variety of automated comparison systems are available using many different facial feature pattern matching algorithms. Typically an image of an unknown offender is used to search a database of 'mug shots' to produce a list of persons that share facial characteristics. The result is then used for intelligence purposes to generate investigative leads and identify potential suspects. The system can also be used to assist in the verification of a suspect’s image against police mug shots. Automated face recognition systems are most reliable in a security setting where there is sufficient control over the quality of the image captured, such as airport screening.
In many jurisdictions digital evidence practitioners are recruited from serving police officers, while in others, practitioners are recruited who have qualifications or backgrounds in the areas of computer science, physics, phonetics, digital imaging signal processing image or communication engineering.