Posted in Forensic Video Analysis

Image Comparison Evidence: The Canadian Experience

Image Comparison Evidence: The Canadian Experience Posted on April 8, 2018

Image comparison evidence has been used frequently in Canadian courts where favorable rulings have been made on the use of such evidence in identifying vehicles, clothing and other objects, and people.  It has been subjected to the same type of challenges as seen in the United States and the more harmful the evidence is to the opposing party the more likely it is to be challenged.

One of the seminal decisions in Canada on the use of image comparison evidence is R. v. Cooper, [2000] B.C.J. No. 446 (British Columbia Supreme Court), where the Court stated:

[90] Leaney and Nikolovski make it clear that a Trial Judge may on his own observations of a video tape, and of his comparisons of the tape to the Accused in the stand, conclude beyond a reasonable doubt that the person seen in the video is the Accused. In Nikolovski, Cory, J., speaking for the majority, reviewed the evolution of the use of audio tapes, photographs, and video tapes as evidence in Canada. At pg. 411 he pointed out that in R. v. B(K.G.) (1993), 1 S.C.R. 740, the Supreme Court of Canada praised the evidence obtained from video tapes as a “milestone” contributing to the “triumph of a principled analysis over a set of ossified judicially created categories”. And after referring to the majority decision of McLachlin, J., as she then was, in Leaney, he had this to say:

Similarly, in R.v. L.(D.O.) (1993) 4 S.C.R. 419, 85 C.C.C. (3d) 289 (S.C.C.) L’Heureux-Dube, J., in concurring reasons, noted that the modern trend has been to admit all relevant and probative evidence and allow the Trier of Fact to determine the weight which should be given to the evidence in order to arrive at a just result. She observed that this is most likely to be achieved when the decision-makers have all the relevant probative information before them. She wrote at pg. 455 that:

“It would seem contrary to the judgments of our Court, (Seaboyer v. B.(K.G.) …) to disallow evidence available through technological advances such as video taping, that may benefit the truth seeking process”.

In my opinion the forensic video evidence of Constable Fredericks did not alter or tamper with the images on the video tape in the case at Bar. His digitization and video analysis was a most useful tool in the performance of my task. It is in my view the type of evidence referred to by Mr. Justice Cory in Nikolovski, and by L’Heureux-Dube, J., in R. v. L. (D.O.), being at most an extension of the video taping evidence.

In R. v. Kalle, 2007 WL 298566 (Ontario Superior Court of Justice), a convenience store clerk was robbed at knifepoint by a masked man.  The robbery was captured on three different surveillance cameras.  The defendant was arrested within two hours of the incident and was wearing a reversible jacket, dark pants and shoes similar to those worn by the robber.  When a search warrant was executed at the defendant’s residence, a black balaclava and a pair of ski gloves were seized.

Part of the evidence for the prosecution was the evidence of a forensic video analyst who was called to give expert evidence regarding the comparison and contrast of known and questioned images.  The analyst testified that he placed the seized items on a mannequin.  Known images were taken and then compared with the questioned video images.  The Court reviewed the detailed comparison and color compensation evidence given by the analyst, noting that the expert evidence served to point out and elucidate the images that were available to the Court.  The Court found the expert evidence to be very helpful though not dispositive of the identification issue before the Court.  It was however an important factor in ultimately convicting the defendant.

In R. v. Mohamed, 2014 ABCA 398 (Alberta Court of Appeal), the trial judge considered evidence on the issue of identification, including the evidence given by a forensic video analyst regarding surveillance video obtained from several sources.  In giving her expert evidence, the analyst testified that while she could not conclusively state that the known suspect was the questioned person seen in the surveillance videos, there were a number of consistencies such that she was able to say that the known suspect was consistent with the questioned person and could not be eliminated.  Relying upon the expert evidence together with other identification evidence, the trial judge concluded that the defendant was in fact the questioned person shown in the surveillance video.  On appeal, the issue was whether the trial judge was entitled, on the available evidence, to reach that conclusion.

The Court of Appeal noted the following regarding the use of the video evidence in deciding the issue of identification:

61     Further, while the expert witness testified to detail not observed by some of the eyewitnesses, nor commented upon by the trial judge, none of this extra detail detracts from the ultimate conclusion that Mr. Mohamed was the shooter. It may be explained by the fact that the expert had the opportunity to slow down the video footage and look at it in the form of still shots. She was able to take the time she needed to make a careful comparison, much more likely to yield small details than the brief time period each of the eyewitnesses had to observe the shooter.

62        Absolute uniformity and consistency in every detail of eyewitness evidence is not only unnecessary, it is highly unlikely in a dynamic situation such as this, where the witnesses were located at different vantage points, with different lighting. This is even more the case where eyewitness impression, gained in the moment, is compared to expert opinion based on time-consuming review of video recordings.

In this case, the forensic video analyst analyzed the images, compared them and gave a conclusion that did not include a positive identification regarding the suspect.  In doing so, she acted in accordance with her training and was cognizant of the amount of individual detail available in the images.

A ruling similar to Mohamed was given in R. v. Mikolajczyk, 2014 CarswellMan 8, 2014 MBCA 3 (Manitoba Court of Appeal).  As discussed above and in the previous post entitled “Image Comparison Evidence: The United States Experience,” when a forensic video analyst conducts an evaluation and comparison of questioned images, especially of people, rarely is the result a conclusive identification.  The role of the expert in such a case is to assist the trier of fact in reaching a conclusion on identification, if possible.  In this case, the trial judge relied on the evidence of the analyst, who pointed out numerous consistencies but did not reach a conclusive opinion on identification, and ultimately concluded that identification had been proven by the prosecution.  

On appeal, the defendant challenged the propriety of the trial judge’s conclusion and the Court of Appeal stated:

3        The trial judge considered all the evidence and found a significant number of similarities between the clothing and the male in the surveillance footage to that of the clothing seized from the accused and his appearance. She found that the “consistencies were numerous and in some instances remarkable” and while, individually, they would have been insufficient to ground a finding of guilt, viewed cumulatively, she found them to be sufficient to be convinced of his guilt beyond a reasonable doubt. While she mentioned some dissimilarities, she pointed out that this could have been a distortion due to the lighting or the video imaging process as explained by the expert.

5        We do not agree that the trial judge speculated as to why some differences existed between the clothing seen on the surveillance footage and the clothing seized from the accused. She accepted the evidence of the expert as to how certain distortions can occur on surveillance footage. We agree with the Crown that there is a difference between speculation and what the trial judge did here, which was to make a reasonable and logical factual inference from the evidence.

6        The trial judge’s detailed reasons make clear that she took into account all of the information available to her. She reviewed the surveillance footage and was able to make her own comparisons between the clothing in the footage and the clothing seized from the accused. She was able to find numerous points of similarity between the various items of clothing and between the physical appearance of the male in the footage and the accused. She also responded in a reasonable fashion to the argument of the accused as to the alleged dissimilarities.

This case is a good illustration of the role of the forensic video analyst in assisting the court in evaluating video evidence and in reaching a supportable conclusion.  When I prosecute cases that involve forensic video analysis evidence, I do not expect my expert to solve the case.  Rather, I expect my expert to provide me with some helpful evidence that I can factor into my final argument before the trier of fact at the end of the case.

There have been a small number of decisions in Canada wherein the court has either restricted the use of image comparison evidence or disallowed it altogether. 

In R. v. Coelen, [1999] B.C.J. No. 2867 (British Columbia Supreme Court), the Court was critical of the comparison slides the forensic video analyst had prepared for use in court.  The Court found the slides to be overly laden with text and graphics and ruled that they were not admissible in the format presented.  

In R. v. Aitken, 2008 CarswellBC 3329 (British Columbia Supreme Court), the Crown sought to tender the evidence of a forensic video analyst who would provide comparison evidence between images of a suspect and known images of the defendant.  While class characteristics would be pointed out for both the person and the clothing, no unique characteristics were noted.  The defense objected to this evidence on the basis that the jury did not require such assistance and that the image overlay evidence was unduly suggestive and therefore prejudicial.  The trial judge ruled that the comparison evidence was inadmissible on the basis that the jury did not require assistance with this task.  Further, the image comparison slides were deemed to be too suggestive and therefore unfairly prejudicial.  The Court ruled that the jury could view all of the images the expert presented but that they would be left to make their own conclusions.  The Court further ruled that the images should be shown without the use of any arrows to point out certain aspects of the images.

The Aitken decision is an outlier as very few courts have ruled comparison evidence to be inadmissible or that the use of arrows and similar focus tools are not permitted.  When properly explained to the court, image comparison evidence should rarely be restricted or ruled inadmissible.

Both the Coelen and Aitken decisions serve as a warning to forensic video analysts.  It is important to keep comparison slides clean and informative but not overpowering.  Excessive use of text and graphics is to be avoided; after all, it should not be a reading test for the trier of fact.  I strongly recommend that when presenting comparison evidence in court, analysts should have two versions prepared and ready to use.  The first version is the one the analyst wants to use, the one that is most informative, using an appropriate amount of text and graphics and is reflective of the work undertaken.  The second version should contain the same images but no text or graphics in the event the court rules that the proposed text and graphics are not admissible.  Using the second version is still helpful as the analyst will explain the details of the comparison to the trier of fact, though without the benefit of additional visual aids. 

 Image comparison evidence is a valuable tool for maximizing the value of video images and in ascertaining the truth.  There is a growing body of case law that supports the use of such evidence.  Given the expertise required to do such work, image comparison should only be undertaken by those with the necessary forensic training.