AI Supporting The Earth
To celebrate Earth Day 2019, we explain how Intel is committed to advancing uses of AI that positively impact our world by providing social good organizations with technologies and expertise to accelerate their work.
By Anna Bethke, Head of AI for social good at Intel
This pale blue dot is all we have. As Carl Sagan said, “There is nowhere else, at least in the near future, to which our species could migrate. Visit, yes. Settle, not yet. Like it or not, for the moment the Earth is where we make our stand.”
As inhabitants of this planet, we face considerable challenges. The most recent United Nation’s Global Environment Outlook warns “urgent action at an unprecedented scale” is needed to protect our global ecosystems.Though the stakes are dire, advanced technology is helping address some of the most critical issues we face. As we celebrate Earth Day here at Intel, I’d like to highlight a few of the artificial intelligence (AI) projects that Intel is working on to help solve environmental problems.
Saving the Whales
Since 2017, Intel has been collaborating with Parley for the Oceans* and a team of marine biologists to collect the mucus exhaled from whales when they surface to breathe, then utilizing AI technology to analyze indicators or the whale’s health in real time. By developing a greater understand what’s threatening their ecosystem, we can better protect whales. Learn more about the Parley* SnotBot and see behind the scenes with the team working on the project.
Moving from the oceans to land, Intel Software Innovator Rosemarie Day is an IoT DevOps engineer and the winner of last fall’s Intel® AI Interplanetary Challenge. Day developed a model to retrieve data from LandSat satellites and combine it with other data sources, like localized weather, to monitor and evaluate deforestation using TensorFlow*. Rosemarie’s work shows how AI can be used to classify plants and analyze deforestation and growth, providing a more informed view of how the world is changing over time. By examining numerous factors, scientists can assess urbanization trends, human health issues, and the magnitude of deforestation over different regions.
Protecting Human Resources
Obtaining clean drinking water can be a difficult and costly problem for many in our world. Peter Ma, another Intel Software Innovator, is using AI and high definition cameras for bacteria classification to ensure safety in water sources. Confirming a clean water source typically requires expensive testing equipment and manual analysis of the results. For regions in the world in which access to clean water is a continuing problem, simpler test methods could dramatically help prevent diseases, save lives, and reduce environmental impact. Started in 2017, the Clean Water AI project has utilized the Intel® Movidius™ Neural Compute Stick to perform water testing in real time with an inexpensive portable system.
Every year, a large number of crops are damaged by pathogen attacks. Farmers suffer huge losses due to severe infestation. To combat this issue and increase yields, proper identification of diseases is the first step toward treatment. Using the Intel® Distribution of OpenVINO™ toolkit, Risab Biswas is creating an optimized type of identifying methodology that will help to identify pathological plant diseases and their proper control measures, thereby helping reduce acreage needed for production.
Sadly, some human actions are much more sinister than the adverse impacts of food and water production. Stopping poachers before they can kill endangered animals is no easy task. Central Africa has lost 64 percent of its elephant population in the last decade. Non-profit RESOLVE’s* new TrailGuard AI* camera uses Intel technology to detect poachers entering Africa’s wildlife reserves and alert park rangers in near real time. TrailGuard AI builds on previously anti-poaching prototypes and uses deep neural network algorithms to allow the device to recognize humans and vehicles with a high degree of accuracy. With the first-generation camera, rangers had to manually review photos to determine if there was a poaching threat or a false-positive triggered by other motion. By adding an additional layer of AI into the system with the tiny yet powerful Intel® Movidius™ Myriad™ 2 VPU, TrailGuard AI now intelligently knows when a potential poacher is present, allowing park rangers to rapidly intercept and apprehend the animal killers.
Supporting Social Good
Intel is committed to advancing uses of AI that positively impact our world by providing mission driven organizations with technologies and expertise to accelerate their work.
Read more on how AI is helping us better understand the environment and check out Intel’s AI4SocialGood page to see additional projects. We are excited to continue to explore additional ways we can help protect this pale blue dot of ours.
Intel, the Intel logo, Xeon, and Movidius are trademarks of Intel Corporation or its subsidiaries in the U.S. and/or other countries. *Other names and brands may be claimed as the property of others. © Intel Corporation
Bio: Anna Bethke's intellectual passions include machine learning, data mining and analytics, GEOINT, and human factors engineering. Anna is particularly interested in how these fields blend together; determining what information may be most valuable to customers and then determining the most intuitive manner to convey these insights. She tweets at @data_beth.
- On-line and web-based: Analytics, Data Mining, Data Science, Machine Learning education
- Software for Analytics, Data Science, Data Mining, and Machine Learning
- An introduction to explainable AI, and why we need it
- AI For Ordinary Folks
- What is missing when AI makes a decision?
- Top Stories, Apr 22-28: The most desired skill in data science; How…
- Data Scientists Have Developed a Faster Way to Reduce Pollution,…
- KDnuggets™ News 19:n16, Apr 24: Data Visualization in Python…
- More Resources for Women in AI, Data Science, and Machine Learning
- Knowledge Graph Forum: Technology Ecosystem and Business Applications
- How to use continual learning in your ML models, June 19 Webinar