The AI of the tiger: A high-tech bid to save an iconic species
It's a storyline worthy of a crime drama. The victims' remains have been discovered, and now it's down to investigators to piece together who they were and where they came from. But the victims aren't people, they're endangered tigers.
The tiger population across Asia has suffered a massive decline over the last 100 years. There are now believed to be fewer than 4,500 wild tigers left in the world. The biggest culprit in the iconic animals' decline has been poaching to feed the illegal trade in tiger parts.
Some of those illegal products are used in traditional medicines, but tiger pelts are also highly prized as decorative items.
The unique stripy coats that make them so sought after, however, are also helping conservation workers tackle the poaching problem — and they should soon get a big helping hand in those efforts from artificial intelligence (AI).
Campaigners have been hard at work building a database of photographs of individual tiger skins. The idea is to be able to identify and track where illegally traded pelts come from, so law enforcement agencies in different countries can be contacted and help shut down wildlife traffickers.
"Every tiger's stripe pattern profile is unique, just like our fingerprints" Debbie Banks, the Tigers & Wildlife Crime campaign leader at the London-based Environmental Investigation Agency (EIA), told French news agency AFP. "So, we can use tiger stripes when we see images of tigers that have been offered for sale online, or if we have images of tigers that have been offered for sale in marketplaces or seized. We can use those images to cross reference against images of captive tigers that might have been farmed for trade and we can help join the dots as to where do the tigers that we see in trade come from."
Poring over thousands of often crowd-sourced photographs of tiger skin rugs, carcasses and even stuffed tigers, Banks and her team at the EIA have been working to match the stripe patterns to individual tigers to track where the animals came from, but its painstakingly slow work.
That, they hope, is about to change.
The EIA was recently awarded one of the first grants from the Alan Turing Institute, a U.K.-based center for data, science and artificial intelligence named after the renowned WWII codebreaker, and it's now developing a tool that uses AI to do the meticulous work of comparison.
"We have a database of images of tigers that have been offered for sale or have been seized, and when new images emerge, when our investigators get new images, we need to scan those against the database. At the moment we are doing that manually, looking at the individual stripe patterns of each new image that we get and cross-referencing it against the ones we have in our database," explained Banks. "The idea is that the artificial intelligence will - basically the scientists will create an algorithm, which means that they identify the individual stripe patterns of a unique individual tiger."
The EIA has appealed to anyone who sees tigers, dead or alive, to submit photographs along with any identifying information available to help build out the database.
"To develop, train and test the technology we need thousands of images of individual tiger stripe patterns, sourced by EIA staff, other organizations and you!" the group said in a news release announcing the project.
"This is a unique opportunity for tiger lovers around the world to get hands-on and actually contribute directly to the future conservation of tigers," Banks said in a statement.
It's hoped the technology will one day be used to help other vulnerable species, and put the people exploiting those species out of business, but for now, the focus is on getting it up and running to help save the big cats.
