From any electric kettle connected to the Internet, you can hear about how AI beats esports, gives new opportunities to old technologies and draws cats from your sketch. But the fact that the machine mind also manages to take care of the environment is less often said. Cloud4Y decided to correct this omission.
Let's talk about the most interesting projects that are being implemented in Africa.
DeepMind tracks Serengeti herds
For the past 10 years, biologists, environmentalists and volunteer conservationists under the Serengeti Lion Research program have been collecting and analyzing data from hundreds of field cameras located in the Serengeti National Park (Tanzania). This is necessary to study the behavior of certain species of animals whose existence is threatened. It took the volunteers a whole year to process the information by examining demographics, movements, and other markers of animal activity. AI DeepMind is already doing this job in 9 months.
DeepMind is a British company that develops artificial intelligence technologies. Alphabet was bought in 2014. Using the data set for training the artificial intelligence model, the scientific team has achieved excellent results: can automatically detect, identify and count African animals in images, doing its job 3 months faster. Why this is important, the DeepMind staff explains:
βThe Serengeti is one of the last remaining places in the world with a pristine community of large mammalsβ¦ As human encroachment around the park intensifies, these species are forced to change their behavior in order to survive. Increasing agriculture, poaching and climatic anomalies are driving changes in animal behavior and population dynamics, but these changes have occurred on spatial and temporal scales that are difficult to control with traditional research methods.β
Why is artificial intelligence more efficient than biological? There are several reasons for this.
- More photos involved. Since installation, the field cameras have captured several hundred million images. Not all of them are easy to recognize, so volunteers have to manually identify the species using a web-based tool called Zooniverse. The database now has 50 different types, but it takes too much time to process the data. As a result, not all photographs are used in the work.
- Rapid species recognition. The company claims that their pre-trained system, which will soon be deployed in the field, is able to perform on par with (or even better than) human annotators, remembering and recognizing more than a hundred animal species native to the region.
- Cheap equipment. AI DeepMind is able to work effectively on "modest" hardware with unreliable Internet access, which is especially true for the African continent, where a powerful computer and fast Internet access can be destructive to wildlife and prohibitively expensive to deploy. Biosecurity and cost savings are important benefits of AI for environmental activists.
It is expected that the DeepMind machine learning system will be able to not only track the behavior and distribution of the population in detail, but also provide data quickly enough so that environmentalists can respond in a timely manner to short-term changes in the behavior of Serengeti animals.
Microsoft is watching the elephants
In fairness, we note that DeepMind is not the only company that has taken care of saving fragile populations of wild animals. So, Microsoft was noted in Santa Cruz with its startup , which uses AI to keep an eye on African savannah elephants.
The Elephant Listening Project startup, with the help of a Cornell University lab, has developed a system that can collect and analyze data from acoustic sensors scattered throughout the Nuabale-Ndoki National Park and adjacent forest areas in the Republic of the Congo. Artificial intelligence recognizes the voice of elephants in the recordings - low-frequency rumbling sounds that they use to communicate with each other, and receives information about the size of the herd and the direction of its movement. According to Conservation Metrics CEO Matthew McCone, artificial intelligence is able to accurately identify individual animals that cannot be seen from the air.
Interestingly, thanks to this project, a machine learning algorithm was developed, trained on Snapshot Serengeti, that can identify, describe and count with an accuracy of 96,6%.
TrailGuard Resolve warns of poachers
An Intel-powered smart camera uses AI to protect endangered African wildlife from poachers. The peculiarity of this system is that it warns of attempts to illegally kill animals in advance.
The park's cameras use an Intel computer vision processor (Movidius Myriad 2) that can detect animals, people and vehicles in real time, allowing park rangers to intercept poachers before they mess things up.
The new technology that Resolve has come up with promises to be more effective than conventional detection sensors. Anti-poaching cameras send out alerts whenever they detect movement, resulting in many false alarms and limiting battery life to four weeks. The TrailGuard camera only uses motion to wake up the camera and only sends alerts when it sees people in the frame. This means that false positives will be much less.
In addition, the Resolve camera consumes virtually no power in standby mode and lasts up to a year and a half without recharging. In other words, park staff won't have to risk their safety as often as they used to. The camera itself is about the size of a pencil, making it less likely for poachers to find it.
What else can you read on the blog?
β
β
β
β
β
Subscribe to our -channel, so as not to miss the next article! We write no more than twice a week and only on business.
Source: habr.com