Identifying Criminals by Hands Visible in Images
Written by Chris Grice with Shane Turnidge   

As a fingerprint examiner, I try to stay abreast of news and events in my field. News articles from around the world relating to my discipline are forwarded to me daily via Google news alerts and I stay involved with a fingerprint interest group where I can openly discuss issues and get feedback from my fingerprint colleagues and peers.

Recently, I began seeing articles about hackers and criminals stealing fingerprints from publicly posted images and using those prints for nefarious purposes. There was extensive coverage of a German hacking group that claimed to have re-created fingerprints of a public official, namely Ursula von der Leyen, the German Minister of Defense, from images taken while speaking to the public. The photographs show her raising her hand with the underside of the hand exposed. Images could be enlarged to capture the detail found in the palm and fingers of her hands. The premise of these articles was that her prints had already been captured and that her identity thus would be ripe for the taking. There were suggestions in the articles that politicians would soon be wearing gloves while speaking in public.

A German hacking group claimed to have re-created fingerprints of German Minister of Defense Ursula von de Leyen from images taken while speaking to the public. There were suggestions in articles that politicians would soon be wearing gloves while speaking in public.

Once you capture someone’s print or prints by this method, what do you do with them? For criminals, I suspect, there is not much they can do. With a public wary of identity theft, the hacker group’s intent is to shock and scare the public, and the publications posting these articles also want to grab attention to gain more Internet views. While it is true that access to secure sites is often facilitated via a finger or thumb scan, especially with new smartphones, the person recreating a print from an image still needs access to that portal at the secure site where the print is scanned. Using fingerprint scans to access money in ATM machines never took off. The potential for fraudulent use of finger and palm prints captured from a public photo is, I feel, very limited. More importantly, as someone who has made their career identifying finger and palm prints, I believe we should turn our thoughts to using this practice as a crime-solving tool.

As forensic scientists, we can turn the table and use advanced technology to fight crime. I am talking about identifying criminals by using friction-ridge skin visible in photographs. I will tell you about a case here in the United States, perhaps one of the first of this kind, in which I did just that. As camera image quality and resolution continues to improve, more and more opportunities to do this will occur.

My case involved two similar sexual assaults of women on different dates in the same rural area in 2008 and 2009. Both victims related that the perpetrator had threatened them using a small, black handgun. After a lengthy investigation, a suspect was developed and an arrest made. A search warrant was executed at the suspect’s residence in hopes of finding the small, black handgun; however, it was not found. A cellphone to which the defendant had access was seized, as one victim had related that the suspect was talking on a cellphone as he fled the assault scene. Investigators checking data on this phone found six close-up images of a human hand holding a small, black handgun. Some of the images showed the backside of the hand, and some the palmar side, including both sides of the weapon. The defendant denied that the hand holding the gun in the images was his. I was approached by an astute investigator from the state attorney’s office, looking to fortify their case, and asked if I might be able to identify the person by the friction-ridge skin visible in the images. One of the pictures had discernible ridge detail at the thenar area of the left palm.

“Hand selfie”: This image was recovered from a cellphone that had been accessible by the defendant. The image depicts the left hand holding a small black handgun. Friction ridge skin from the thenar area of the palm is visible in the image. This friction ridge skin was identified to the defendant, who is also believed to be the photographer. Photo credit: Evidence Image

I have to admit, when I was first asked to look at this case, I was skeptical that the resolution would be good enough for identification purposes. I believe that at that time, the top-of-the-line cellphones had only 3 megapixel cameras. I felt that the more I enlarged the small area of friction ridge skin in the image, the more the resolution would be lost. In this case, it was likely that the photographer was holding the gun in one hand while holding the camera in the other to take the pictures, so the camera was likely within a few feet of the friction-ridge skin. The defendant’s known palmprint impressions were on file. Once I began to make the side-by-side comparison however, the level 1 detail (ridge flow and direction) and level 2 detail (gross ridge paths or characteristics, creases) all fell into agreement and my confidence in doing this grew. The ridge edges are not crisp in the image but the resolution is sufficient.

A cropped and enhanced view isolating the thenar area of the palm, taken from the “hand selfie” image on Page 8, showing detail of the friction ridge skin. Photo credit: Evidence Image


A known left palm impression of the defendant (cropped from a live-scan card) with the thenar area of the palm in the red boxed-off area. Photo credit: C. Grice

(left) A mirror image or horizontal flip of the known thenar palm detail of the defendant for the comparison chart. (right) A tonal or color reversal of the known thenar palm detail of the defendant for the comparison chart. Photo credit: C. Grice

I have always felt strongly that with any identification, I must be able to convince the jury that the conclusion was correct. When I testified at trial in this case, I explained that normally we compare a questioned impression to a known impression to determine if both impressions are made by the same source. In this case however, the friction ridge skin in the photograph is a questioned source, not an impression, being compared to a known impression. Impressions are mirror images of the source item. Just as a person with a scar on his left cheek would see that scar on the right side of his image while looking in a mirror, detail and ridge flow in an impression will be the same but symmetrically reversed of what is observed in the source friction-ridge skin that made that impression. It was important for the jury to understand why I had done a horizontal flip of the known palm impression photographically to compare it with the questioned friction-ridge skin. I used an analogy in court that this is similar to comparing the sole tread design on a shoe or sneaker to a footwear impression in snow, mud or ink, to determine if the questioned footwear made the imprint. I then took my shoe off my foot in court, held it up to the jury and showed them the brand name lettered into the sole tread. I showed them that the brand name could easily be read left to right, when viewing the sole, but if I were to make an impression in mud or snow with my shoe, the brand name would be viewed backwards in the impression. I further went on to explain that the summits of the friction ridges in the image of the hand holding the gun were reflecting the sunlight and appear light, whereas the furrows are in shaded valleys and are dark. This is why I made a photographic tonal reversal of the recorded impression so that I could compare tonally light colored ridges to light colored ridges and dark furrows to dark furrows. I opted to do the mirror image flip and tonal reversal on the known impression rather than the questioned friction ridge skin as I felt it was best not to alter in any way the questioned detail. As examiners, these concepts are familiar, but to the lay person, it is so important that they “get” these issues. When it came to the part where I was showing the ridge events in agreement, the jury was seeing it for themselves and nodding in agreement. Visual aids also allow the jury to stay focused and conceptualize the issue at hand better than solely relying on verbal testimony.

The charting of matching detail with the questioned friction ridge skin on the left and the known impression on the right. The detail in agreement is similarly numbered in both prints with red lines denoting locations. Photo credit: C. Grice / J. Brunetti

The random capture of friction ridge skin in images that are of comparison quality is happening more and more via the advances in cameras and electronic image resolution technology. The popularity of social media sites such as Facebook, Instagram, and Twitter, to name just a few, are creating a plethora of publicly accessible photographs. It is likely we will see an increase in these types of identifications in criminal investigations if the investigators are informed of what can be accomplished.

A colleague in Canada, Shane Turnidge, reports he has had three such cases in the last three years. His cases involve images used to identify gang members and a similar gun-related case. As two of his cases have not yet been adjudicated in court, he is only able to share with us one such case. Turnidge shared with me the particulars of this case as follows:

Redacted screenshot of an image acquired from a Twitter account. The image depicts a left hand holding bullets. The friction ridge skin in the image was identified to the defendant, who is also believed to be the photographer. Photo credit: Evidence Photo

A digital image was observed by an investigator in Canada who was keeping tabs on social media postings by known gang members and affiliates. He knew the subject in question was prohibited from being in possession of firearms and explosives as part of his parole. The investigator became curious when the image of the palm of a left hand holding bullets appeared on one of the subject’s social media accounts. This investigator has had previous experience with Turnidge on another image found on a cellphone and reached out for an opinion on the suitability of this image for identification.

When Shane completed his analysis of the image, he advised the investigator that the image had significant value and that it was indeed suitable for both comparison and identification purposes.

The investigator then forwarded Turnidge the identity of a potential suspect and asked him to compare his recorded prints with the information available within the image from the social media account. Before he could compare the photograph to the inked palmprint impression, Turnidge realized that, like myself, he would need to synchronize the friction ridges in the photograph with the known palmprint impression record. To do that, he chose to horizontally flip the image so that the palm image would be consistent with the impression of a left palmprint. As the friction ridges appeared light in the photograph and the friction ridges appeared dark in the palm impression, he chose to invert the photograph, essentially creating a photographic negative in which the friction ridges appeared dark. The photographic image was now suitable for comparison purposes to the known left palmprint record.

The image from Twitter was horizontally flipped, inverted, and rotated. Photo credit: S. Turnidge

This image depicts the charting of the unknown image (cropped) with the known left palm impression of the known subject showing some of the level 2 features in agreement between the two images. Photo credit: S. Turnidge

As a result of the comparison between the image of the unknown person holding the bullets and the known palmprint record of the subject, an identification of the hand in the image was made to the suspect provided by the investigator.

Turnidge relates that sizing an image to a known record can be challenging. In this particular case, he decided to use Photoshop to calibrate and size the evidentiary photograph to become approximately the same size and image resolution as the known palmprint record. Both images were then magnified to make the demonstrative chart.

Turnidge and I both feel there is merit in trying to get the message out to law enforcement to be aware that this evidence can be very useful in any type of occurrence. Police investigating cases of child abuse and exploitation should be especially alert to the use of this investigative technique, as the potential of friction-ridge skin being visible in images from these types of cases is great.

There are, of course, challenges to identifying individuals by their friction ridges visible in images. One such challenge is that the questioned image and the known impression have not been calibrated to scale. In my case, I was able to ascertain the dimensions of the handgun visible in the image from manufacturing specs, and thus, to some extent, calibrate size of the questioned hand with the known impression. Even absent this, however, the identification process involves the physical relationship of the friction-ridge characteristics to one another (i.e. amount of intervening ridges), not the measured distance between those characteristics. The finger impression of a small child, and an impression from the same finger after that child has grown to adulthood, can easily be identified as coming from the same source, despite differences in size. Turnidge and I both agree it should not be an issue of significance.

Secondly, Turnidge points out that resolution issues can be equated with clarity values in any friction-ridge evidence. The higher the resolution and the better the focus, the more small intrinsic features you should be able to resolve.

Each image will present its own challenges. Turnidge further suggests that any examiner working with this type of evidence should be cognitive of any potential situational or contextual bias that might influence their findings. This is negated when the focus of the examiner is on the friction-ridge evidence and not on other aspects of the image.

In closing, the final photo from my case depicts the back of the hand holding the gun. As forensic scientists, we should also be cognizant of the biological uniqueness demonstrated in this photo as well. There are knuckle creases and scars visible that could be compared, if only we make the effort to get the known standards for comparison. Although there is very little case law for skin-crease identification, let’s change that. Similarities in two things being compared can and should be documented whether or not individualization is being considered.

Here is another image captured from the same cellphone that had been accessible by the defendant in the first case discussed in this article. The image depicts the back of a hand (knuckle side) holding a small black handgun. How is biological uniqueness demonstrated in this photo? Photo credit: Evidence Image

Finally, a digital-savvy co-worker, John Brunetti, helped me considerably with my case. Of additional significance, Brunetti was able to determine that the cellphone images in my case had metadata that included GPS location information identifying where the images were taken, the result of a geo-cache setting found on many smartphones. The pictures were found to have been taken in close proximity to the sexual assault sites.

There is an old saying that “A picture is worth a thousand words.” I say that this estimate is an insult to images everywhere! They are worth so much more if only we take the time and effort to see.

Editor’s Note

As we readied this article for press, the New York Daily News published an article on June 23, 2015 about how detectives at the Sarasota County Sheriff’s Office utilized fingerprints visible in child pornography images to identify—and eventually convict—the offender. If you know of any other cases similar to this one and the examples in this article, the authors would like to hear about them.Their contact information is listed below.

About the Authors

This e-mail address is being protected from spam bots, you need JavaScript enabled to view it is an International Association for Identification Certified Latent Print Examiner and Senior Crime Scene Analyst employed with the Division of Scientific Services of the State of Connecticut in the Forensic Science Laboratory.

This e-mail address is being protected from spam bots, you need JavaScript enabled to view it is a Latent Print Examiner employed by the Peel Regional Police in Ontario, Canada.

< Prev   Next >

Lifting Latent Fingerprints from Difficult Surfaces

ALMOST ANYONE can find, process, and lift a latent print that happens to be in a logical and obvious place like a door handle, a beer can, or a butcher knife. But sometimes, a latent print is not just sitting there in a logical and obvious place. Sometimes, you have to use your imagination to find the print and your skills to lift it.