Evidence exists in many forms, especially within the digital world. It’s common for cases to have an unidentifiable person in a video or image. Similarly, law enforcement may have another photo to confirm this individual’s identity but often can’t reach any conclusions independently. Consequently, digital forensic analysis is vital in these instances. At Primeau Forensics, we offer forensic image comparison services to help you uncover the truth.


Forensic image or video comparison is the process of comparing known and unknown subjects depicted in digital imagery. Subsequently, we identify whether an unknown subject matches the known subject. At Primeau Forensics, we offer image comparison, video comparison and other identification services. These services assist criminal defense cases, law enforcement applications and litigation uses. In addition to these services, we offer a detailed analysis based on case similarities and differences.

Learn how our comparison services solved a real case of mistaken identity, and explore the forensic guidelines we adhere to below.

Forensic comparison begins with an assessment between imagery features. This assessment helps to identify or eliminate subjects. For example, photographic comparisons include:

  1. A facial comparison between an unknown subject depicted in a surveillance image with an identified suspect.
  2. The comparison of objects, such as a vehicle, depicted in surveillance images with those recovered in an investigation.

Guidelines for Forensic Image Analysis are set by the Scientific Working Group on Digital Evidence (SWGDE).


A digital image comparison or video comparison investigation is bound by images of sufficient quality. A wide variety of digital camera systems and formats can create these images. As a result, sufficient image quality is necessary to analyze and extract characteristics for photo forensics analysis.

The most prolific source of police evidence is video images. However, video is not an accurate medium for identification on its own. This is because digital compression removes and adds information that wasn’t originally included. And so, an expert must have the ability to interpret data to provide an accurate comparison.

When trying to qualify images, we examine qualities such as:

  • Frame rate.
  • Resolution.
  • Compression artifacts.
  • Motion blur.
  • Lighting.

If our experts feel that they have a clear enough picture and data to analyze an image, they can move to the next step.




Sufficient for Identification in Forensic Image Comparison


After qualifying all images in question, we begin our forensic image comparison. This is a two-step process. First, a thorough examination of class and individual characteristics between the known and unknown allows an analyst to form an opinion. Second, if the objects or subjects in question are the same, the analyst will eliminate all others of the same class. 

You can explore the distinctions between class characteristics and individual characteristics further below. Standards for comparison criteria are set by SWGDE and LEVA.


These are identifiable features that narrow the likelihood that a known and unknown object or subject belongs to the same group. However, keep in mind that consistent class characteristics between objects and subjects in question can only suggest similarities. That being the case, they cannot infer a positive identification on their own. Possible class characteristics could include a clothing item from a certain brand or a shoe size. While these characteristics may help an expert narrow the suspects down to a smaller group, they cannot guarantee identity.


These are observable features that distinguish objects within a class from one another. Individual characteristics result from a variety of conditions. These conditions include natural and manufacturing processes, intentional alteration, and normal wear-and-tear.

The ability to identify a specific object or subject requires similarity between individual characteristics. With this in mind, analysts determine the sufficiency of these characteristics based on quality, quantity, and persistence of imagery details.

Possible individual characteristics could include handwriting features and fingerprints. Since these qualities differ from person to person, they're more likely to help an expert reach a reasonable conclusion. However, these character types are less likely to appear than class characteristics.


The experts at Primeau Forensics utilize a scale commonly known as the “likelihood scale” or the “scale of conclusions.” The scale describes the strength of our opinions coupled with the comparison investigation. Though this confidence scale originated in the United Kingdom, experts now use it on an international level.

The visual quality of the evidence, present characteristics and undefined differences all play a role in making conclusions. Explore the scale in the accordion below.

The imagery in question must be of good quality. At this level, we can compare most general or limited class characteristics in detail. And it is likely that these characteristics are evident in multiple images.

When observed in combination with one another, the available imagery allows for robust comparison. Then we can make a distinction between two items of clothing, objects, or vehicles of very similar appearance. In addition, there are no significant unexplained differences.

There are many general class and limited class characteristics that we are able to compare in detail at this level. And it is likely that the characteristics are evident in multiple images depending on the image quality.

Individually or in combination with one another, the available imagery allows for robust comparison. This should allow us to make a distinction between two items of clothing, objects, or vehicles of similar appearance. In addition, there are no significant unexplained differences.

There are multiple class characteristics that we are able to compare in detail at this level. And it is likely that some limited class characteristics are evident in multiple images. However, we may only observe them on one occasion depending on the image quality.

We may or may not observe these characteristics in combination. In addition, there are no significant unexplained differences.

The image quality at this level permits us to observe a moderate amount of object detail. For example, several visible class characteristics described individually or in combination. We may find observable limited class characteristics. In addition, there are no significant unexplained differences.

We are able to observe a limited number of class characteristics at this level. For example, a class characteristic is a dark-toned long sleeve top.

Individually, these features only allow for a superficial comparison. Therefore, we have limited ability to discriminate between two objects. In addition, there are no significant unexplained differences.

Due to the poor quality, we are not able to observe any visible similarities at this level. Therefore, there is not enough support for an expert to make a more meaningful comparison than a layperson.

It is unsafe to offer any support towards the contention that the two items under comparison are one and the same. However, this does not exclude the objects from being one and the same.

There are significant unexplained differences between the two items at this level. Therefore, this excludes the items from being one and the same.


Expert witness services are available for image comparison investigations by Primeau Forensics. Our expert witness services include a forensic report of all our testing, plus trial preparation and testimony. Primeau Forensics can generate a formalized report outlining our process, comparison methodology, and opinions. This assists the trier of fact to make determinations about identification of subjects, objects, or vehicles.

When we’re preparing for an appearance in court, we work with the attorney to decide the best questions to ask us about our investigation. This process helps the attorney understand the situation and present the facts in the best way possible.

When we prepare for testimony, we anticipate all possible direct and cross-examination questions. Furthermore, we will involve experts from other industries should it strengthen our investigation. This will, in turn, strengthen our own reports and testimony. 

Our forensic experts are trained to handle the expert witness position. We strive to remain unbiased in litigation and offer a third-party perspective for the evidence at hand.

Digital Image Forensics and Expert Witness Services