{"id":2515,"date":"2022-08-22T14:00:46","date_gmt":"2022-08-22T14:00:46","guid":{"rendered":"https:\/\/salarydistribution.com\/machine-learning\/2022\/08\/22\/an-ai-enabled-drone-could-soon-become-every-rhino-poachers-horn-enemy\/"},"modified":"2022-08-22T14:00:46","modified_gmt":"2022-08-22T14:00:46","slug":"an-ai-enabled-drone-could-soon-become-every-rhino-poachers-horn-enemy","status":"publish","type":"post","link":"https:\/\/salarydistribution.com\/machine-learning\/2022\/08\/22\/an-ai-enabled-drone-could-soon-become-every-rhino-poachers-horn-enemy\/","title":{"rendered":"An AI-Enabled Drone Could Soon Become Every Rhino Poacher\u2019s\u2026 Horn Enemy"},"content":{"rendered":"
\n

Want inspiration? Try being charged by a two-ton African black rhino.<\/p>\n

Early in her career, wildlife biologist Zoe Jewell and her team came across a mother rhino and her calf and carefully moved closer to get a better look.<\/p>\n

The protective mother rhino charged, chasing Jewell across the dusty savannah. Eventually, Jewell got a flimsy thorn bush between herself and the rhino. Her heart was racing.<\/p>\n

\u201cI thought to myself, \u2018There has to be a better way,\u2019\u201d she said.<\/p>\n

In the latest example of how researchers like Jewell are using the latest technologies to track animals less invasively, a team of researchers has proposed harnessing high-flying AI-equipped drones powered by the NVIDIA Jetson edge AI platform to track the endangered black rhino through the wilds of Namibia.<\/p>\n

In a paper published this month in the journal PeerJ<\/a>, the researchers show the potential of drone-based AI to identify animals in even the remotest areas and provide real-time updates on their status from the air.<\/p>\n

For more, read the full paper at https:\/\/peerj.com\/articles\/13779\/<\/i><\/b>.<\/p>\n

While drones \u2014 and technology of just about every kind \u2014 have been harnessed to track African wildlife, the proposal promises to help gamekeepers move faster to protect rhinos and other megafauna from poachers.<\/p>\n

\u201cWe have to be able to stay one step ahead,\u201d said Jewell, co-founder of WildTrack<\/a>, a global network of biologists and conservationists dedicated to non-invasive wildlife monitoring techniques.<\/p>\n

Jewell, president and co-founder of WildTrack, has a B.Sc. in Zoology\/Physiology, an M.Sc in Medical Parasitology from the London School of Tropical Medicine and Hygiene and a veterinary medical degree from Cambridge University. She has long sought to find less invasive ways to track, and protect, endangered species, such as the African black rhino. <\/span><\/p>\n

In addition to Jewell, the paper\u2019s authors include conservation biology and data science specialists at UC Berkeley, the University of G\u00f6ttingen in Germany, Namibia\u2019s Kuzikus Wildlife Reserve and Duke University.<\/p>\n

The stakes are high.<\/p>\n

African megafauna have become icons, even as global biodiversity declines.<\/p>\n

\u201cOnly 5,500 black rhinos stand between this magnificent species, which preceded humans on earth by millions of years, and extinction,\u201d Jewell says.<\/p>\n

That\u2019s made them bigger targets for poachers, who sell rhino horns and elephant tusks for huge sums, the paper\u2019s authors report. Rhino horns, for example, reportedly go for as much as $65,000 per kilogram.<\/p>\n

To disrupt poaching, wildlife managers must deploy effective protection measures.<\/p>\n

This, in turn, depends on getting reliable data fast.<\/p>\n

The challenge: many current monitoring technologies are invasive, expensive or impractical.<\/p>\n

Satellite monitoring is a potential tool for the biggest animals \u2014 such as elephants. But detecting smaller species requires higher resolution imaging.<\/p>\n

And the traditional practice of capturing rhinos, attaching a radio collar to the animals and then releasing them can be stressful for humans and rhinos.<\/p>\n

\n

\"\"

Above it All:<\/strong> Observing rhinos from above leaves the animals undisturbed, while letting friendly humans know of any threats IMAGE CREDIT:<\/strong> WildTrack.<\/figcaption><\/figure>\n

It\u2019s even been found to depress the fertility of captured rhinos.<\/p>\n

High-flying drones are already being used to study wildlife unobtrusively.<\/p>\n

But rhinos most often live in areas with poor wireless networks, so drones can\u2019t stream images back in real-time.<\/p>\n

As a result, images have to be downloaded when drones return to researchers, who then have to comb through images looking to identify the beasts.<\/p>\n

Identifying rhinos instantly onboard a drone and alerting authorities before it lands would ensure a speedy response to poachers.<\/p>\n

\u201cYou can get a notification out and deploy units to where those animals are straight away,\u201d Jewell said. \u201cYou could even protect these animals at night using heat signatures.\u201d<\/p>\n

To do this, the paper\u2019s authors propose using an NVIDIA Jetson Xavier NX<\/a> module onboard a Parrot Anafi drone.<\/p>\n

The drone can connect to the relatively poor-quality wireless networks available in areas where rhinos live and deliver notifications whenever the target species are spotted.<\/p>\n

To build the drone\u2019s AI, the researchers used a YOLOv5l6 object-detection architecture. They trained it to identify a bounding box for one of five objects of interest in a video frame.<\/p>\n

Most of the images used for training were gathered in Namibia\u2019s Kuzikus Wildlife Reserve<\/a>, an area of roughly 100 square kilometers on the edge of the Kalahari desert.<\/p>\n

\n

\"\"

Mother knows beast (mode):<\/strong> African black rhinos are known to be protective of their young. IMAGE CREDIT:<\/strong> WildTrack.<\/figcaption><\/figure>\n

With tourists gone, Jewell reports that her colleagues in Namibia had plenty of time to gather training images for the AI.<\/p>\n

The researchers used several technologies to optimize performance and overcome the challenge of small animals in the data.<\/p>\n

These techniques included images of other species in the AI\u2019s training data, emulating field conditions with many animals.<\/p>\n

They used data augmentation techniques, such as generative adversarial networks<\/a>, to train the AI on synthetic data, the paper\u2019s authors wrote.<\/p>\n

And they also trained the model on a dataset with many kinds of terrain and images taken from different angles and lighting conditions.<\/p>\n

Looking at footage of rhinos gathered in the wild, the AI correctly identified black rhinos \u2014 the study\u2019s primary target \u2014 81 percent of the time and giraffes 83 percent of the time, they reported.<\/p>\n

The next step: putting this system to work in the wild, where wildlife conversationalists are already deploying everything from cameras to radio collars to track rhinos.<\/p>\n

Many of the techniques combine the latest technology with ancient practices.<\/p>\n

\"\"<\/p>\n

Jewell and WildTrack co-founder Sky Alibhai have already created a system, FIT, that uses sophisticated new techniques to analyze animal tracks (see image of a rhino track, left). The software, initially developed using morphometrics \u2014 or the quantitative analysis of an animal\u2019s form \u2014 on JMP statistical analysis software, now uses the latest AI techniques.<\/p>\n

Jewell says that modern science and the ancient art of tracking are much more alike than you might think.<\/p>\n

\u201c\u2019When you follow a footprint, you\u2019re really recreating the origins of science that shaped humanity,\u201d Jewell said. \u201cYou\u2019re deciding who made that footprint, and you\u2019re following a trail to see if you\u2019re correct.\u201d<\/p>\n

Jewell and her colleagues are now working to take their work another step forward, to use drones to identify rhino trails in the environment.<\/p>\n

\u201cWithout even seeing them on the ground we\u2019ll be able to create a map of where they\u2019re going and interacting with each other to help us understand how to best protect them,\u201d Jewell says.<\/p>\n

All Images courtesy of WildTrack<\/b><\/p>\n

\u00a0<\/p>\n

\u00a0<\/p>\n

\u00a0<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

https:\/\/blogs.nvidia.com\/blog\/2022\/08\/22\/ai-drone-rhino-poachers\/<\/p>\n","protected":false},"author":0,"featured_media":2516,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[3],"tags":[],"_links":{"self":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/2515"}],"collection":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/comments?post=2515"}],"version-history":[{"count":0,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/posts\/2515\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media\/2516"}],"wp:attachment":[{"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/media?parent=2515"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/categories?post=2515"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/salarydistribution.com\/machine-learning\/wp-json\/wp\/v2\/tags?post=2515"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}