Search Results for: facial recognition

Lip-tattoo identification is the prevailing standard for authenticating that each horse arriving at the paddock for the next race is not there by mistake–or is not a ringer.

The science of biometrics now offers a contemporary approach to perform this task in a noninvasive way. The promising methodology is facial recognition software, which is increasingly being deployed by a variety of organizations to verify human and animal identity.

In the aftermath of the terrorist attacks of September 11, 2001, some airports installed facial recognition software. At that time, it was susceptible to being fooled; for example, if a person wore glasses or performed a facial contortion.

Since then, the accuracy of facial recognition software has been improved dramatically owing to 3-d scanning and the capacity to map unique features like scars and skin pores.

In the summer of 2011, Facebook introduced facial recognition software that identifies photos of individuals without their names attached. Other notable adopters of the technology include the military, police departments, casinos, and the FBI. When Navy Seals dispatched Osama bin Laden, facial recognition software was used to assist in corroborating his identity.

Recent tests by the National Institute of Standards found that the best facial recognition software can correctly identify two photographs of the same person more accurately than humans can, or over 99 percent of the time. One company sells an employee identification machine for $745 that it says can distinguish between pictures of identical twins.

The usefulness of facial recognition software is not limited to humans. To illustrate, the technology has made it much easier for game wardens to track wild animals like apes and elephants so their behavior can be studied.

Science Daily reports that apes are now routinely videotaped while feeding, in trees, and traversing the forest. To the naked eye, it is difficult to distinguish one ape from another, even in proximity. But facial recognition software has made it possible to sort out apes from their pictures, with a growing degree of validity.

Facial recognition software would be much more exact in confirming the identities of racehorses arriving at a paddock before a race than it is in identifying apes in their natural habitat. Each horse’s face would be photographed up close and then the image would be instantaneously compared to a registered photo of the animal. This protocol is precisely how employees are often cleared to enter secured workplaces.

The Jockey Club and racetracks need to experiment with a combination of facial recognition software and scannable radio frequency identification chips implanted under a horse’s skin–with an eye toward abandoning the crude and decidedly low-tech lip tattoo.

Copyright © 2012 Horse Racing Business

Originally published in the Blood-Horse. Used with permission.

In 2012, I wrote an article for Blood-Horse magazine pertaining to the potential use of facial recognition software in identifying racehorses (click here for the article).  After it was published, I received an email from the founder of a Thoroughbred aftercare facility wanting to know how the technology might be employed to ascertain the names of horses in danger of being sent to slaughter.  Sometimes, lip tatoos can be difficult to read and in other instances a full-blooded Thoroughbred may never have been registered.  And most other horse breeds don’t use lip tatoos. 

A website called Finding Rover demonstrates the opportunity for applying facial recognition software for locating at-risk horses.  A dog or cat owner uploads a picture of their lost pet to the Find Rover website.  Similarly, animal shelters and Finding Rover users upload pictures of found dogs or cats.  Facial recognition software is used to find matches–and then owners are notified. 

The facial recognition software developed for Finding Rover is 99 percent accurate.  The main obstacle that Finding Rover must overcome is not accuracy, but rather is having enough pet owners and animal shelters participate.

A website like Finding Rover for horses would take a while to get up and running.  It would require a database of horse photos and widespread cooperation from aftercare facilities.  Whether a Finding Rover type of website for horses would work or not is an open question.  However, it is worth a try, given the number of animals that are sent to slaughterhouses every year.  Maybe Finding Rover could be persuaded to expand beyond dogs and cats to include horses. The folks at the Thoroughbred Aftercare Alliance should explore the possibilities. 

Copyright © 2018 Horse Racing Business