When wildlife conservationists Zoe Jewell and Sky Alibhai were visiting Zimbabwe to study black rhino populations in the late 90s, they discovered a major issue with how they were tracking the animals.
By tranquilizing the rhinos and fitting a radio-enhanced collar around their neck, the pair were able to study the rhinos’ migration patterns easily. But after years of data generation and analysis, it became obvious that female rhinos were producing fewer children due to the tranquilizing and collaring technique.
Keeping Conservation Conservative
This technique is very common to track animals and their day-to-day, season-to-season habits, but the signs pointed to possible disruptions in the population from this small human intervention.
When the pair began working with the Shona tribe in Zimbabwe, the indigenous trackers were true experts in figuring out information from animal tracks without ever laying eyes on the animals. The amount of information they were able to deduce was astounding: it included species, weight, and sex.
Jewell recalls how the trackers would “often laugh at us as we were listening to these signals coming from the collars. They would say to us, ‘All you need to do is look on the ground.”
The team wondered if the indigenous trackers’ techniques could be programmed into an AI application that would non-invasively keep an eye on any animal population. Over time, it became clear that the answer is a resounding “yes!”.
The pair founded WildTrack in 2004, a non-profit focused on developing a footprint identification technique (FIT) and software that other researchers can use in tracking animal populations. Collaborating with the Shona tribe, a new AI-enhanced drone technology was born. The FIT technology has a 95% accuracy rate, and it’s 100% non-invasive for animals.
Using non-invasive technology like the kind that WildTrack developed helps keep well-intentioned conservation efforts free from human interference and disturbances. Stuart Pimm is a professor of conservation ecology at Duke University. He says that being “collared” creates a lot of stress and anxiety for the animal, and “if you … woke up with something around your neck, I think you’d be in pretty bad shape too.”
Making a Mark Across the World
Jewell claims that WildTrack’s technique can even figure out how many animals are in the wild and where they usually hang out. And even though the FIT technology was created with lots of help from the indigenous trackers, the technology is still not as good as the Shona trackers. Jewell says, “It’s humbling in a way to look at what these expert [trackers] can do in the field. We can only still emulate a tiny bit of that.”
WildTrack’s FIT technology and software are being utilized by dozens of scientists in conservation projects all around the world.
In Namibia, rhinos are monitored; in Portugal, researchers are watching Eurasian otters. In Nepal, tigers are being tracked, while researchers in China are no longer combing through panda feces to track them. About 5 hours outside of Dallas, Texas researchers are tracking mountain lions.
The FIT technology is faster, more accurate, and easier to use.
Although Pimm was not a fan of the original collaring technique, he’s very excited about the FIT technology; “it’s technologically very clever, and it feeds directly into good natural history,” Pimm says.
Another approach that WildTrack has taken is to ask the public to submit footprint images. The non-profit even dedicated a page on their website to teaching users how to take great footprint photos. Now the organization receives hundreds of new footprint images every week from every corner of the world. As a result, the focus for the researchers has shifted to finding better computing hardware and technology (like AI) to increase analysis speeds.
In 2018, WildTrack worked with software development company SAS to create a machine learning application that flags footprints in images and creates marker points in the footprint image. Prior to 2018, this process was excruciatingly manual, but the system has improved the speed of the FIT process by 5 times.
More Than One Application
WildTrack is also applying AI to its footprint-gathering technique. Collaborating with senseFly, a European drone company, the two organizations are building an AI-enhanced drone to track footprints and generate information about the animals using drones. Eventually, the two teams want to create a drone that can scan footprints in higher image resolution and follow animal tracks through grasslands and deep snow.
WildTrack’s software has garnered attention from the U.S. government, who is using it to help park rangers stop animal poaching and illicit trafficking of rhino horns and elephant tusks. The government is also looking at possibilities in using this technology to track humans.
As for Jewell, she’s happy that the U.S. government and WildTrack’s goals align so well to help conserve endangered animals, protect them from harm, and put an end to illicit animal parts trading.
It’s interesting that software originally built with the help of indigenous trackers is being applied by a major government in a variety of other ways. For conservationists, these new applications can only improve the WildTrack software that they love using.
The Future of Animal Protection
Protecting and tracking animals is ultimately done for their benefit, and humans should remain as non-invasive as possible in helping these animals thrive in the wild. With WildTrack, the future is here today. Pimm’s all for the conservative approach to conservation.
He says, “Non-invasive monitoring, I think it’s very much an idea that’s time has come. It’s going to be the wave of the future.”
What do you think about this unique AI application? Do you think there is room for innovation in this field with other emerging technologies, such as IoT? Let us know your thoughts in the comments!Tags: AI and machine learning, AI and ML, AI App Developer, AI app development Dallas, AI drones, app development Dallas, benefits of machine learning, conservation with AI, Dallas AI app developers, Dallas app developers, Dallas machine learning app development, Dallas mobile app developers, machine learning, machine learning app developer, machine learning app developer Dallas, mobile app developers Dallas, wildlife conservation