In order for video evidence to be admitted at trial, the tendering party must prove that the images are authentic.  There are a number of technical approaches that are available to assist with video authentication.  This post will address the use of audit trails as part of the process of establishing authentication.

Proper forensic practice should include the use of audit trails wherein the analyst documents the history of the images from the time the analyst receives them until the forensic work is completed.  This trail would include the documentation of all processes used to enhance or clarify the images.  Some forensic software programs automatically record such steps.  Additionally, some CCTV systems produce electronic audit trails for images captured by the system.  Thus, it is possible to have an audit trail that records everything that happens to a digital video image from the moment of capture until its presentation in court.

The United Kingdom House of Lords Select Committee on Science and Technology noted the following regarding audit trails:

Audit trail

Audit trail methodology is already used with other forms of evidence to ensure that it has been treated fairly, that any processing it has undergone (eg in a forensic laboratory) is well documented, that its location is always known and that all those who have had contact with it have the necessary authorisation. This is all common sense and similar controls would be expected for digital images. If an image is generated by a system dedicated to the enforcement of law then it is possible to start the audit trail at the moment of image capture, but if an image is from elsewhere, for example a photograph from a member of the public or even a CCTV video from a private surveillance system operator, then the audit trail may only start once the image is presented (or seized) by which time it may have already been modified. The courts deal with such problems with all evidence, but the ease with which digital images can be modified, the availability of the technology and the skills to do this, and the low awareness of the potential problem, should raise extra concerns.

In 2002, the United Kingdom Home Office Scientific Development Branch issued a very helpful written set of guidelines (SOPs), entitled Digital Imaging Procedure (Version 1.0, March 2002),[i] to assist law enforcement agencies and forensic video analysts when dealing with digital video evidence.  These guidelines were updated and expanded in 2007 (Digital Imaging Procedure (Version 2.1, November 2007))[ii] and  deal with a number of important topics.  On the issue of audit trails, the following is set out:

One of the fundamental requirements of digital imaging is the need to safeguard the integrity of images; part of this process involves an audit trail being started at the earliest stage.  This may be a written audit trail, and/or incorporate an auto-generated electronic audit trail mapping the movement and changes of files on computers.This Procedure relies on the written audit of activities.  Where good practice is in place for the collection of evidence, including video and still images, there will be no change in principle.  In practice, there probably will be little change in existing procedures with conventional photography except that the operator may receive reusable media to reformat and use; a process familiar to video operators.

The audit trail should include the following information (with date and time of action) when available and if appropriate:

• Details of the case.
• GPMS classification of the image (and any special handling instructions, if relevant) and the name of the person who classified the image.
• If the image is third-party generated, information about point of transfer including whether the image is the Master copy, a Working copy or an exhibit derived from a Working copy.
• Information about capture equipment and/or hardware and software used, including details of the maintenance log relating to capture equipment and calibration of hardware and software.
• Identity of the capture operative including third parties and image retrieval officers, where applicable.
• Details of the exhibits and disclosure officer(s).
• Details of the images captured, including sequencing.
• Details of retrieval or seizure process and point of transfer, if applicable.
• Creation and definition of the Master copy and associated metadata.
• Storage of the Master copy.
• Any access to the Master copy.
• Viewing of the Master and Working copies, including a record of any associated viewing logs.
• Details and reasons for any selective capture.
• Any editing applications which may alter the image.
• Any details of processing applications allowing replication by a comparatively trained individual.
• Electronic history log of processing applications.
• Any copying required to ensure longevity of the data.
• Revelation to the CPS of the Master and Working copies.
• Any copying carried out as part of a migration strategy to ensure the replay longevity of the image.
• Disposal details and retention time periods.

Audit trails have attracted limited judicial attention in North America.  In United States v. Seifert, 351 F. Supp. 2d 926 (D. Minn. 2005), the United States District Court in Minnesota dealt with a case in which part of the evidence against the defendant was surveillance video.  The defendant challenged, among other aspects of the video evidence, authentication.  The analyst performed the following processes on the video images:

• digitized the original analog tape
• slowed it down to real time
• edited the video to show only relevant images
• circled images of interest for the purpose of identifying the arsonist
• enlarged the circled images
• adjusted the brightness and contrast of the circled images
• The defendant objected to the latter three steps, arguing that these processes made the evidence untrustworthy and therefore inadmissible.

On the issue of image authentication and the use of audit trails, the Court commented at page 927:

The witness testified that he made the enhancements using a commercially available software program. He said a newer program not only performs these functions, but also records the keystrokes used to do so. Had the newer software been used, another person could attempt to replicate the work, which would allow more acute analysis of any purported technical flaws in the enhanced images. The Court does not find that the use of the older software taints the enhanced images or bars admission of the enhanced tape. The Court does, however, suggest — by way of dicta — that technology which provides a digital trail could provide an even stronger forensic basis for admission of enhanced electronic evidence.

The defendant argued that the digitized and enhanced copy was no longer an accurate duplication of the original analog images.  In analyzing this issue, the Court said, at page 928:

Defendant claims the enhanced video no longer accurately records the surveillance images, and, as such, is no longer an admissible original recording. The Court does not agree, and finds the proffered tape may be admitted.

The government has carried its burden and laid adequate foundation showing the enhanced tape “accurately reproduce[s] the scenes that took place, [and is] . . . accurate, authentic and trustworthy.” United States v. Beeler, 62 F. Supp. 2d 136, 148-49 (D. Maine 1999). During the evidentiary hearing, the Court viewed the “before” and “after” images. The Court finds the enhanced version to be a fair and accurate depiction of the original videotaped image. The enhancements more readily reveal, but remain true to, the recorded events. As such, they may be entered into evidence before the jury.

The first image transfer — from analog to digital format — changes the image only in a metaphysical sense. The viewer’s perceived image is identical. This transition has no effect on the accuracy of the image. In the Court’s view, it is an “equivalent [duplication] technique which accurately reproduces the original.” Fed. R. Evid. 1001 (4).

Beyond this, the Court finds that adjustments to brightness or contrast, or enlargement of the image, while arguably a manipulation, are in fact no more manipulative than the recording process itself. The image is black and white; the world is not. In the non-digital world, a camera’s lens, its aperture, shutter speed, length of exposure, film grain, and development process — all affect the image. Each of these is entirely unremarkable so long as the “image” remains an accurate recording of that which occurred before the camera. If a photographic negative were magnified by lens, and an enlarged image resulted, no one would question the larger picture. Similarly, in the event of a tape recording, no one would comment if the volume were increased to make a recorded conversation more easily heard — again, so long as the volume-increased words were accurately recorded by the recording medium.

Here, the evidence showed that the technician adjusted the digital image’s brightness and contrast, but maintained the relationships between the light and dark areas of the image. As an example, if the moving figure’s clothing were a certain percentage lighter or darker than the wall behind it, that light/ dark relationship was maintained. The technician testified that he simply “moved” the brightness relationship on the scale, increasing the light’s intensity while maintaining the image’s integrity.

As a result, the Court finds that the enhanced tape accurately presents a true and accurate replica of the image recorded by the Co-Operative’s security camera. It does so in a fashion which maintains the image while assisting the jury in perceiving and understanding the recorded event. Compare  Beeler, 62. F. Supp. 2d at 149 (admitting videotapes where enhancements omitted “extraneous frames” and “the images are larger, clearer and easier to view”); United States v. Luma, 240 F. Supp. 2d 358, 368 (D. V.I. 2002) (admitting videotapes where enhancements “did not change the substance of the videotape, but merely clarified the tapes.”)

Accordingly, the Court finds the enhanced videotape is a duplicate admissible under the best evidence rule. Fed. R. Evid. 1001 (4), 1002, and 1003.

Accordingly, the defendant’s motion to exclude the enhanced video images was denied and the prosecution was permitted to use the images at trial.

This ruling was appealed to the United States Court of Appeals for the Eight Circuit, 445 F.3d 1043 (8^th Cir. 2006).  The defendant argued that the District Court erred in admitting the enhanced video because no one could be certain whether it accurately reflected the original video given that the State’s forensic video analyst did not record the changes made to the original video.  On this point the Court of Appeals ruled that since the analyst described in detail each step taken in the enhancement process, that was sufficient to establish authentication, even though he did not keep copies of the images at each stage of the enhancement.  The Court of Appeals was further satisfied that adjustments to contrast and brightness were applied to all images, not just selected portions.  Accordingly, the District Court was correct in allowing the admission of the video evidence at trial.

United States v. Seifert is a helpful case in endorsing the value of audit trials.  While an audit trail is not a legal requirement in establishing authentication, it is an important component of good forensic practice and worthy of consideration by all agencies that are involved in forensic video analysis.

[i] United Kingdom Home Office, Police Scientific Development Branch, Digital imaging procedure, version 1.0, March 2002; http://www.homeoffice.gov.uk/docs/digimpro.pdf

[ii] http://scienceandresearch.homeoffice.gov.uk/hosdb/publications/video-and-imaging/58-07_-_DIPv2.0.pdf?view=Binary