Dear friends,
This time we have a very exciting news to share with you: We have recently kicked-off of our next project – Marine Litter Detection via Satellites!
This project is about creating a Computer Vision system that is capable of identifying marine litter in the oceans from satellite images in (almost) real-time. The outcome from this project will be accessible for researchers, environmentalists, activists, authorities, and other organizations for no charge.
Over 70% of our planet is covered by water and over 95% of the water bodies are seas and oceans. Marine litter is a major ecological threat to marine ecosystems. It can trap and kill marine life, smother their habitats, pose as hazards to navigation, and also lead to loss of biodiversity. Because of the plastic decomposition into the microplastic, tiny toxic particles can also be ingested by aquatic life causing the toxic particles move up the food chain, and finally threaten the health of humans.
Therefore, cleaning the oceans and preventing further pollution of the oceans are extremely important for our planet and our health. To enable this, it is essential to know where marine litters are located. Currently, location of marine litter is mainly found via local surveys and on-ground/sea observations. This approach is expensive, and time-consuming, not scalable, and leads to relatively sparse data resulting in only rough estimations of the overall amount of marine litter.
In our Marine Litter Detection project, we will create a Computer Vision system that can help improve the current status quo and will allow improved fact-based strategic decision making and better coordination of ocean cleaning efforts. In addition, we plan to present the output of this system in form of a publicly available interactive map to help further research of this problem as well as to increase public awareness of this topic.
This project is a very special one for MI4People for multiple reasons: First, it is our first project that focuses on environmental protection – a topic that is becoming more and more important every day. Second, it will be run not by volunteers alone but together with a German Data Science Consulting agency Alexander Thamm! This great company has enabled this project by providing its highly skilled and motivated employees Arne Hartz, Anastasia Heide, and Leonard Rosen who are working with us pro bono! MI4People gives large thanks to all of them for their exceptional dedication and commitment! And a very special thanks goes to Dr. Johannes Nagele and Wolfgang Reuter from Alexander Thamm who helped initiate this great collaboration! You can find more details about the project under: https://www.mi4people.org/marinelitterdetectionviasatellites
And all these great achievements would not be possible without you and all our other supporters! Therefore, on behalf of all the MI4People volunteers, we would like to say you a big THANK YOU! 😊
In the rest of the newsletter, and as we do almost every month, we have collected some interesting examples for you about how Machine Intelligence (MI) can be used for the benefit of Public Good – in this issue we will focus on how MI is helping (or can help) to protect marine environments and species. So, happy reading this newsletter, put your capabilities to help Public Good delivery into action, and let us together make the world a better place for all of us!
Your MI4People Team
From the World Around Us
AI/ML helping with a deep understanding of ocean’s environment quality
As the oceans are a vital part of our planet, their well-being has a direct impact on us. So, better we understand the ocean’s environment the better we can protect them from a multitude of man-made stressors that harm the oceans. According to Ocean Assessment, time is of the essence in managing these stressors and saving the ocean environment.
The Lofoten-Vesterålen Ocean Observatory (LoVe Ocean) of Norway is very active in the ‘save the ocean’ cause and has been collecting a lot of data. LoVe Ocean and the Institute of Marine Research (IMR) partnered with Capgemini to gain insights from large volumes of data. Capgemini, in turn, used the strength of AI/ML and the AWS cloud platform to create a solution that can sift through very large datasets fast and efficiently thus enabling monitoring, managing, and researching the underwater ocean environments.
Some of the outcomes of this LoVe Ocean and IMR work will be provided to the Norwegian Marine Data Center for broader distribution. The UN is also a party to this initiative, hence, insights from this effort will help international policies around sustainable ocean economy. For further reading, see this article series.
A smart color-based analysis of ocean health by “OC-SMART”
A special type of ocean data, its color, can be a useful indicator of its health. The Ocean Color - Simultaneous Marine and Aerosol Retrieval Tool (OC-SMART) was created for analysis of data obtained by satellite ocean color sensors and is helping researchers efficiently collect information about the ocean’s health on a global scale. This machine-learning platform analyzes the color of coastal and open water to determine the ocean’s health. The data OC-SMART collects looks at factors like pollution and chlorophyll concentrations around the world for a comprehensive understanding of how healthy ocean waters are.
Detection of plastic pollution in the rivers of Asia using AI/ML
As we mentioned in the editorial of this newsletter, plastic pollution in water bodies has many severe harmful effects. The CounterMEASURE project of the United Nations Environment Programme (UNEP), with support from the Geoinformatics Center (GIC) at the Asian Institute of Technology, is using AI/ML to analyze a combination of geospatial data and volunteer-supplied images of plastic pollution to assess the size and impact of plastic pollution in the Mekong River. Such analyses are critical to the understanding of plastic pollution in the oceans drained by the rivers into them. Here AI/ML is also empowering ‘citizen science’ focused on the well-being of our planet.
Speeding up ocean cleanup using AI/ML
A Dutch NPO called The Ocean Cleanup develops technology-based sustainable solutions of plastic pollution clean up of oceans as well as for detecting and intercepting plastic pollutants in the rivers before they can reach the ocean. These solutions use a lot of image analyses which have significant manual component making these solutions tedious and slow. Using a set of Microsoft’s cloud-based technologies from the Microsoft Azure platform, including Azure Machine Learning cloud services, these manual processes for image recognition have been turned into (nearly) automatic ones thus speeding of the overall cleanup process significantly and in a cost-efficient way.
Concluding Remarks
As we have seen in the various examples cited in this newsletter, AI/ML technologies have advanced sufficiently and in the context of the right use cases they can be of great help to us in doing our jobs much better. In this newsletter, we focused on the topic of the health of marine environment as it is critical for the broader sustainability and biodiversity protection issues. In previous newsletters we have shown AI/ML applications in several other Public Good areas. Additionally, you will find more examples and related discussions at the MI4People website.