Intel has recently partnered with Accenture and the Sulubaaï Environmental Foundation to create an AI-driven data collection platform aimed at analyzing and protecting vulnerable marine habitats, habitats like coral reefs.
A combination of climate change, pollution, and overfishing have been damaging the world’s oceans, particularly coral reefs. Coral reefs around the world are experiencing mass die-offs and problems like coral bleaching. Scientists and conservationists are looking for ways to protect coral reefs and help them recover. Designing plans to support coral reefs requires data, and as Engadget reported, Intel has partnered with two environmental foundations to create the CORaiL platform. The purpose of CORaiL will be collecting information on coral reefs and other marine habitats, providing researchers with the data they need to determine what strategies could be effective at protecting vulnerable marine ecosystems. As Jason Michell, managing director of the Communications, Media, and Technology practice at Accenture explained in a blog post:
“Artificial intelligence provides unprecedented opportunities to solve some of society’s most vexing problems. Our ecosystem of corporate and social partners for this ‘AI for social good’ project proves that there is strength in numbers to make a positive environmental impact.”
In May of last year, the team of researchers and engineers from the three organizations installed concrete structures along reefs found near the Philippines’ Pangatalan Island. The concrete chunks contained sections of living coral capable of growing into new habitat for creatures inhabiting coral ecosystems. In addition, the researchers placed video cameras underwater near the structures so they could collect data on the coral and the surrounding environment. The cameras utilized an AI-driven video analytics system developed by Accenture, and the cameras enabled the researchers to gather data on the reefs through minimally invasive methods.
Accenture’s AI video analytics system lets researchers collect real-time video data from the coral environments, without needing to be physically present in the water. While many divers collect footage of coral reefs, this incurs travel expenses and presents the possibility that the divers could interfere with wildlife in the area. The AI video platform does much of the data collection and analysis for the research teams, continually monitoring the environment for change, and letting researchers do analysis in more or less real-time.
Over the course of the past year, CORaiL has collected around 40,000 images for analysis, and the images are already helping researchers analyze how coral reefs change in response to shifting environmental conditions. Meanwhile, engineers from the cooperative effort are already working on the next generation of the CORaiL system. The next proptype will include a backup power supply and an optimized series of convolutional neural networks. New versions of CORaiL might be employed for tasks other than studying coral, such as studying how tropical fish migrate through cold waters or monitoring for violators of reef protection orders.
CORaiL isn’t the only new project to make use of AI with the goal of protecting the oceans. A new AI system designed by researchers from the Plymouth Marine Laboratory in the UK tracks plastic pollution in the ocean through the analysis of satellite imagery. The AI system analyzes imagery collected by the European Space Agency’s (ESA) satellites and finds large chunks of floating debris by analyzing the “spectral signature” produced by the trash (patterns of light absorbed and reflected by the trash). After training, the AI was able to recognize a multitude of different objects when tested on images of seas from Vietnam, Canada, Ghana, and Scotland. The AI reportedly achieved approximately 86% accuracy when differentiating trash from natural objects.
According to the scientists involved in the research, their experiment marks the first time that plastic pollution has been tracked with satellites. The research team wants to improve the technique and enable it to detect patches of trash within rivers and along coastal regions.
AI Models Help Predict Large Tropical Waves and Ocean Currents
A team of researchers has recently designed an AI model that is capable of forecasting oceanic phenomena that stretch over hundreds of miles/kilometers, like tropical instability waves (TIW). Tropical Instability Waves (TIW) are an oceanic event that takes place within the Pacific Ocean, near the equator. The Pacific TIW involves the motion of curved, triangular waves that move westward along the edges of the tropic pacific cold tongue – a region of the tropics notably colder than the ocean surrounding it.
The environmental factors that give rise to TIW are extraordinarily complex and the phenomenon is hard to forecast. Forecasting the TIW is traditionally done with complex statistical and physical models. However, a team of researchers has recently designed an AI model intended to better predict TIWs and other ocean phenomena.
According to Phys.org, the research team was headed by the Institute of Oceanology of the Chinese Academy of Sciences (IOCAS) Professor LI Xiaofeng, and the team also included members from various Chinese oceanic sciences divisions like Shanghai Ocean University and the Second Institute of Oceanology of Ministry of Natural Resources. The team made of use satellite data to design a deep learning model intended to analyze the instability waves as they move thousands of kilometers through the ocean. Even with global satellite data, the environmental factors that impact oceanic phenomena can be difficult to discern, but the goal is that AI models can do a better job of deciphering these variables and making predictions than traditional models.
The deep learning model that the researchers designed made use of sea surface temperature data collected by satellites, analyzing current patterns and contrasting them with surface temperature data collected in years past. The researchers had the model trained and tested on approximately nine years of data. When the results were analyzed, the researchers found that the model was able to accurately and consistently forecast the change in sea surface temperature and predict temporal and spatial variations within TIW as a result.
The study implies that AI models supported by big datasets can be reliable ways of forecasting even some of the most complex phenomena in the oceans.
“AI-based models, statistical models, and traditional numerical models can complement each other and provide a novel perspective for studying complicated oceanic features,” LI Xiaofeng explained according to Phys.
It’s hoped that as the model is improved and refined it will assist in the prediction of large waves and storms, which has practical applications for ocean navigation systems and the prediction of severe weather events that may harm coastal cities. This type of research is particularly valuable in a world where climate change is altering how ocean currents move and interact with the environment around them. The research conducted by LI Xiaofeng and colleagues is part of a growing trend of using AI algorithms and satellite data to learn about and predict the movement of ocean currents and related phenomena.
As another example of AI being used to track and predict oceanic phenomena, earlier this year a team of researchers from the Plymouth Marine Laboratory and the University of the Aegean published a study examining how machine learning algorithms and satellite data can be used to identify areas of concentrated plastic waste and track its spread.
The team took satellite images of plastic refuse and trained an image identification system on them, aiming to see if the system could accurately discern patches of plastic trash from wood, seaweed, and other natural floating objects. According to the results of the study, the algorithm was able to correctly identify garbage approximately 86% of the time. The researchers want to improve upon the model and create a system that could facilitate the identification and cleanup of plastic waste along coastlines and rivers.
AI Models to Help Identify Invasive Species of Plants Across the UK
Environmental scientists and artificial intelligence researchers are utilizing AI to fight an invasive species spreading across the UK. Researchers from the UK Centre for Ecology and Hydrology (UKCEH) and Birmingham have developed an AI model intended to survey regions like roadsides for the presence of various invasive species, including Japanese knotweed.
Japanese knotweed is an invasive species that can do damage to natural landscapes and buildings around the UK, as it’s able to damage the foundations of buildings. It’s often considered one of the most damaging and aggressive invasive plant species in the UK. Getting rid of Japanese knotweed often proves challenging because it proves challenging to find and identify. AI researchers are hoping that the machine learning algorithms can cut down on the time and resources needed to identify Japanese knotweed.
Training data was collected for the model through the use of high-speed cameras placed on top of vehicles, which collected images of approximately 120 miles of vegetation on the roadside. Ecologists will examine the images and label the knotweed, and the images will have their GPS location tagged. The labeled images will then be used to train a computer vision model to recognize samples of Japanse knotweed. The same process will be used to recognize other species of invasive plants found in the UK, such as Himalayan balsam and rhododendrons. The system will also be used to detect ash trees, which are native to the UK but are at risk of being decimated by disease.
The AI model will be tested over the course of a 10-month pilot project. The research team says that there are challenges that the team needs to overcome, such as being sure that the images captured by the cameras are of consistent quality and that when there are multiple species in a single image all species are properly identified. If the pilot program ends up delivering promising results, it could end up being adapted for use in other countries around the globe, helping these countries battle their own invasive species problems. As computational ecologist at UKCEH, Dr. Tom August, was quoted by The Next Web:
“Invasive plant species tend to grow in corridors, which is why we’re focused on roadside surveys a computational ecologist at UKCEH. If the pilot is successful, this could be scaled up in other countries, or for other species of plants, trees or even insects and animals.”
According to August, AI models open up many possibilities for learning about the natural world and engineering efficient, cost-effective solutions to invasive species. UKCEH is collaborating with Keen AI, an AI company based in Birmingham. The founder of Keen AI, Amjad Karim, was quoted by Science Focus as saying that the use of AI models to analyze images and detect invasive species can help reduce costs and provide safety to landowners, highway agencies, and policymakers. The primary method of gathering roadside images currently requires surveyors, and that road is temporarily closed while they complete their work.
The new project designed by UKCEH and Keen AI is just the latest in a growing trend that sees the application of AI to fight invasive species. Late last year, AI researchers from Microsoft and CSIRO joined forces to design an AI model that can an invasive species called para grass, found throughout Kakadu National Park in Australia. Para grass is a fast-growing weed that can spread rapidly, quickly displacing many native plants in a region. The researchers utilized images collected by drones, and once the model was trained on the labeled images it was able to successfully identify para grass, allowing the researchers to remove it from vulnerable wetlands. This had the effect of allowing thousands of magpie geese to return to the region. Yet another team of researchers from the New University of Alberta used machine learning models to design containment and mitigation strategies for various invasive species in Canada.
Intel & Accenture Discuss Using AI to Save Coral Reefs – Interview Series
We sat down (virtually) with Patrick Dorsey, the Vice President of Product Marketing, Programmable Solutions Group, Intel and Jason Mitchell, a managing director in Accenture’s Communications, Media & Technology practice and the company’s client lead for Intel.
We discussed how on Earth Day 2020, Accenture, Intel and the Sulubaaï Environmental Foundation decided to partner to use artificial intelligence (AI) – powered solution to monitor, characterize and analyze coral reef resiliency in a new collaborative project called CORail.
On Earth Day 2020, project CORaiL was announced, what was it about this project that caused you to take notice?
Jason Mitchell: Coral reefs are some of the world’s most diverse ecosystems, with more than eight hundred species of corals building and providing habitats and shelter for approximately 25% of global marine life. The reefs also benefit humans — protecting coastlines from tropical storms, providing food and income for 1 billion people, and generating US$9.6 billion in tourism and recreation annually. But reefs are being endangered and rapidly degraded by overfishing, bottom trawling, warming temperatures and unsustainable coastal development. This project allowed Accenture and our ecosystem partners to apply intelligence to the preservation and rebuilding of this precious ecology and measure our success in a non-intrusive way.
Could you describe some of the technology at Intel that is being used in the underwater video cameras?
Patrick Dorsey: The underwater cameras are equipped with the Accenture Applied Intelligence Video Analytics Services Platform (VASP) to detect and photograph fish as they pass. VASP uses AI to count and classify the marine life, with the data then sent to a surface dashboard, where it provides analytics and trends to researchers in real time, enabling them to make data-driven decisions to protect the coral reef. Accenture’s VASP solution is powered by Intel® Xeon® processors, Intel® FPGA Programmable Acceleration Cards, an Intel® Movidius™ VPU and the Intel® Distribution of OpenVINO™ toolkit.
Work is currently being undertaken on the next-generation CORaiL prototype. What advanced features will this prototype have compared to the current version of CORaiL?
Jason Mitchell: We are scaling our work in the Philippines with a next-gen Project: CORaiL prototype, which will include an optimized convolutional neural network and a backup power supply. We are also looking into infra-red cameras which will enable videos at night to create a complete picture of the coral ecosystem. These technology advances will allow our solution to scale to look at new use cases like: studying the migration rate of tropical fish to colder countries and monitoring intrusion in protected or restricted underwater areas.
Could you share some of the computer vision challenges that are involved in monitoring different fish populations in an underwater setting which may result in significant changes in lighting conditions?
Patrick Dorsey: A critical element of Project: CORaiL is to identify the number and variety of fish around a reef, which serve as an important indicator of overall reef health. Traditional coral reef monitoring efforts involve human divers manually capturing video footage and photos of the reef, which is dangerous and time-intensive and can disrupt marine life, as divers might inadvertently frighten fish into hiding.
CORaiL monitors coral reef health in the Philippines, are there plans on expanding to other regions?
Jason Mitchell: It’s still early days with this technology, so we’re currently focused on the reef surrounding the Pangatalan Island in the Philippines.
Is there anything else that you would like to share about CORaiL?
Jason Mitchell: AI should be an added contributor to how people perform their work, rather than a backstop for automation. For Project: CORaiL, AI is empowering our engineers to achieve more and learn faster when it comes to growing the coral reef. It empowers the solution to gather data in a non-intrusive manner, allowing the scientists and data engineers to gather data from the reef with minimal disruption to this fragile ecology.
What are some of the other ways AI is being used for Social Good?
Patrick Dorsey: At Intel, we are working with partners to use AI to curb anti-poaching of endangered animals, to map vulnerable populations, to help the quadriplegic community regain mobility and more. We are deeply committed to advancing uses of AI that most positively impact the world.
Jason Mitchell: Through our Responsible AI practice at Accenture, we help organizations implement governance frameworks and tools to ensure they’re deploying AI in a way that aligns to their corporate values and mitigates unintended consequences.
- Dimitris Vassos, CEO, Co-founder, and Chief Architect of Omilia – Interview Series
- Human Brain’s Light Processing Ability Could Lead to Better Robotic Sensing
- Game Developers Look To Voice AI For New Creative Opportunities
- Udacity Launches RPA Developer Nanodegree Program in Conjunction with UiPath
- AI Used To Identify Gene Activation Sequences and Find Disease-Causing Genes