The integration of Artificial Intelligence and the Internet of Things has the potential to aid in monitoring marine plastic waste as well as the biodiversity of marine ecosystems, including the abundance of animal populations.
AI-assisted annotation helps users in marine biodiversity observations from real-time video feeds. Image Credit: Wave Labs
Many measurement stations globally furnish air quality data, facilitating its enhancement. However, accessing data from marine areas proves notably challenging due to poor signal transmission underwater, hindrances from pressure differentials and currents on measurement devices, and the lack of pre-existing computing infrastructure.
Professor of Computer Science Petteri Nurmi and the research team at the University of Helsinki have teamed up with researchers from the University of Tartu, University of Madeira, and MARE-Madeira, ARDITI, a not-for-profit marine research institute, to develop solutions that integrate sensor technologies and embedded Artificial Intelligence, to improve marine data collection. The researchers aim to enhance the data collection methods currently used in environmental research, making them more efficient and widespread.
The higher the quantity and quality of data about the oceans obtained, the better we can use it to understand and protect the oceans. Our methods help expand the total amount of data gathered from marine areas and reduce the effort required to collect and analyze them.
Petteri Nurmi, Professor, Department of Computer Science, University of Helsinki
AI Identifies Animal Species
Nurmi and colleagues conducted a study that utilized data collected from whale-watching tours in Madeira, Portugal. These tours typically have individuals on board who can document observations of the species they encounter or record footage of the environment during the excursion.
During the study, an AI model was employed to assist individuals in real-time observations of the environment. The AI model was also used to identify specific animals, such as whales or dolphins, in the recorded footage. Furthermore, the study examined the effectiveness of AI assistance for both experienced and less experienced observers, as well as how AI-assisted animal observations could be used to train the model.
Nurmi explained“We analyzed how AI assistance affected the quality of data and human observations. AI improved the accuracy of animal observations by amateurs but had no effect on expert observations. On the other hand, when the data collected were used for training purposes, the best results were achieved by combining AI classifications with expert observations. Thus, interactions between humans and AI can influence each other and they need to be better understood.”
The technique has the potential to facilitate the prompt identification of marine animals in motion. Moreover, the findings and approaches can be extended to monitor other life forms.
Identifying Marine Plastics
A recent study was conducted to identify and classify plastic debris present underwater, which is different from microplastics as it is visible to the naked eye. The conventional method of collecting samples through divers or devices for laboratory analysis is time-consuming. Aerial photography is used to observe surface-layer plastics.
The researchers developed an AI model that uses sensors and analyzes light spectrum data to determine the type of plastic waste present underwater. The model can be connected to diver equipment or a diving robot. They found that the model was able to differentiate between different types of plastic with an accuracy of 85%.
We are capable of identifying four out of five objects directly, which means we need to send fewer samples to a lab for identification. This provides us with more data and thus a more comprehensive overview of marine plastic pollution.
Petteri Nurmi, Professor, Department of Computer Science, University of Helsinki
Nurmi and colleagues aim to innovate new approaches to gather data. Other experts can then utilize these methods to their best advantage. For example, identifying the type of plastic can facilitate the determination of its origin and ways to prevent pollution. These data also aid researchers in comprehending how different plastic types impact ecosystems since they possess distinct chemicals and undergo decomposition in various ways.
Better Data for Better Protection
Nurmi thinks that the state of marine data might resemble the situation that existed in air quality research some years ago.
Nurmi added, “Originally, air quality research relied on a few large measuring towers, but now you can find them even on bus stops. In marine sciences, this is not yet the case, even if the methods for collecting marine data and the number of actors involved are increasing. Our research ensures that we can obtain even greater amount of even more accurate marine measurement data as the capabilities to collect data improve.”
The collection of marine data is expected to witness a rise in the usage of underwater drones, smart buoys, vessels, and coastal base stations in the future. The growing network of such equipment will present additional means for obtaining underwater data.
Despite their vital importance for humanity, marine protection and regulation, unfortunately, tend to be drowned out by other issues. The higher the quantity and quality of marine data collected, the better we can develop solutions and regulations improving the status of the marine environment.
Petteri Nurmi, Professor, Department of Computer Science, University of Helsinki
Journal Reference:
Radeta, M., et al. (2024) Man and the Machine: Effects of AI-assisted Human Labeling on Interactive Annotation of Real-Time Video Streams. ACM Transactions on Interactive Intelligent Systems. https://doi.org/10.1145/3649457.
Source: https://www.helsinki.fi/en