On a world populated by sentient anthropoid species evolved from Terran felines, both big and small cats, could the local authorities use the fur patterns and colors the same way that we used fingerprints for decades to identify people?

Or, once their technology reaches the same level than we have today (2018), through pattern recognition software via cameras much as we have facial recognition software (See China for the extreme example of massive population surveillance via cameras on the streets)?

Would that be a viable way of identifying specific persons in the population?

  • 4
    $\begingroup$ In human forensics, fingerprints are typically used to match an individual's identity after the fact. I'm not sure fur patterns even could serve that same purpose, since they aren't really left behind. I do remember reading about at least one species of feline where the fur patterns are as distinct as human fingerprints, but that doesn't help you if there's nothing to compare to... $\endgroup$ – a CVn Nov 24 '18 at 7:42
  • 1
    $\begingroup$ Maybe. But even if it can't actually be done, that's not what's important in fiction. It's reasonable and that's what's important in fiction. $\endgroup$ – RonJohn Nov 24 '18 at 8:06
  • 2
    $\begingroup$ see this site: leopardidproject.com/about.html "We use techniques to ID different animals, for example: spot patterns, Knicks, cuts, eye colour, size - amongst others. And we aim to eventually use facial and spot pattern recognition software." $\endgroup$ – Real Subtle Nov 24 '18 at 11:21

I live somewhere where, a year ago, sadly, an enormous complex of fires ravaged neighborhoods over 3 counties. Although the flames didn't quite reach my town, I evacuated with my 6 cats. I then closely followed the work of the multiple groups of people who spent months rescuing and trapping cats in and near the burned areas (some even ongoing).

These folks came up with methods of cataloging the cats they found. There were many hundreds of cats. A good number were from pre-existing feral colonies or were strays that had been abandoned before the fires. But a lot were animals separated from their humans because of the fires.

Another issue is that most of these animals were filthy (the ash in the air and covering every outdoor (and indoor) surface for miles was insane, and hard to wash off) and some had been injured in ways that made it difficult to recognize them (often because of bandages in medical facilities). But, even for cats that more or less looked the same as they did pre-fires, identification was difficult.

The rescuers took pictures of each cat they found and had records indicating the medical or shelter facility the cat was taken to. If the cat was found by a regular person, the facility would create the records. (This system was not perfect and many people would just care for cats themselves.)

The primary way they classified the cats was by color. While many cats were put into more than one color category, a more detailed system, like the one your story would use, would account for the variations and have more choices. So instead of "tabby," you might have "gray tabby" and "gold tabby." And you'd have groupings for cats of more than one color.

You might think the classification of fur length (short, medium, long) would be primary, but I think the physical issues resulting from the fires made that impractical. Of course fur length was noted in descriptions. Along with gender, intact status, size, age, and anything else of note.

And this brings us to markings. A lot of cats (none of mine, but most of the fire cats) did have markings of some sort. Many desperate people searched for their pets using online resources and info on found pets got posted to several websites and Facebook groups.

Using the pattern of the markings became an art form. There was no good nomenclature for it, but people worked endlessly to match pictures of lost cats with pictures of found ones (dogs were easier to ID but a lot of the lost animals were livestock, where there were similar issues to cats).

I saw so many heart wrenching cases where a distraught person would post a picture of her/his cat and a dozen people would say "I just saw a picture of this cat in the found section!" but it would be the same, wrong, cat the person had seen a dozen times. Sometimes the pictures of these unrelated cats would be so close that only the owner could tell the difference, and only in person (which they did, over and over again).

In one case that made the news, a shelter prematurely adopted out a fire cat and the new family refused to give back the cat when the real family came forward (losing your house is traumatic and it took them a few weeks to track down the cat). The markings from photographs matched the cat but it wasn't enough to prove identity. Authorities finally used DNA. It matched and the adoptive family was compelled to return the cat to its family.

What's the takeaway? Color and markings can change over time. With age, with cleanliness, and with injury. The only reliable method for matching a cat with its human family turned out to be microchipping (and that failed much of the time because people move and don't update their contact info...a lot of the found cats had their names in their listings, but no way to find their families).

I've now chipped all my cats and it's the only method I could see for a high tech society to identify cats who walk by a camera. Even for ID done when booking a crime suspect, for example, I don't think markings will work in many cases, because they change over time. But that will depend a lot on the way your cats are marked...if everyone has prominent markings, clear enough to be spots and stripes and etc, then maybe. If it's like real-life cats, it won't be good enough.

  • 1
    $\begingroup$ This article doesn't go into detail about how they classified cats and matched them with their families, but it has some background. In case anyone is interested (I am putting it in a comment since it doesn't really answer the question). pressdemocrat.com/lifestyle/8385864-181/… $\endgroup$ – Cyn Nov 24 '18 at 19:46
  • 1
    $\begingroup$ The 'with injury' part is important, and can have some pretty striking visual effects. For most domestic cat breeds, if the skin itself suffers significant deep injuries like burns, frostbite, or severe allergic reactions, the fur in that area will usually only grow back white, regardless of the cat's normal fur color. $\endgroup$ – Austin Hemmelgarn Nov 25 '18 at 18:27
  • $\begingroup$ @AustinHemmelgarn That happens with people too! $\endgroup$ – Ynneadwraith Nov 26 '18 at 13:40
  • $\begingroup$ @Ynneadwraith Huh, I actually did not know that. The only reason I knew about it happening in cats though is my black cat who has some random white patches from injuries he got as a kitten before he got picked up by the shelter I adopted him from. $\endgroup$ – Austin Hemmelgarn Nov 26 '18 at 15:15

Yes it will work barring disguise or face/body covering.

Tigers have striped skin not just striped fur. The stripes are like fingerprints and no two tigers have the same pattern.


Bonus fact

A cat's fur and skin colors are closely related. https://pets.thenest.com/skin-pigmentation-cats-11322.html

Because of the above, disguises would only be superficial. They could dye their fur but the skin underneath would preserve the pattern. This wouldn't need shaving - just parting the fur. Not much use with cameras at a distance though.


Yes, but you wouldn’t want to use it on its own.

If the aim is to be able to recognise humans in real time from a video feed then fingerprints are of absolutely no use. You need facial recognition or something similar.

If the pattern of hairs is distinct enough for each cat that they can distinguish between each other using them then you can absolutely train a computer to use fur pattern recognition, but the failure rate could be quite high (fur blurs, people might take countermeasures like using dyes or wearing clothes). If you couple it with something like gait recognition though (tracking how people walk, which is pretty unique from person to person) and feature distance recognition (shoulder to hip, hip to leg, length of arms, position of eyes etc) then you should be able to get pretty good matches.


Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.