What is it about?

This work introduces SALINA, a system that uses sonar technology to monitor wildlife and ecosystems in remote areas. Sonar works by detecting objects using sound waves, and this system makes it possible to track animals such as fish and otters in real-time. Designed to operate in challenging environments with limited power and internet access, the system uses solar energy and advanced data processing techniques to ensure continuous monitoring. By helping researchers and conservationists gather critical information about wildlife behavior and populations, this system supports efforts to protect ecosystems and manage natural resources.

Featured Image

Why is it important?

SALINA is a novel sonar analytics system uniquely designed for deployment in remote, off-grid ecosystems, where traditional monitoring technologies fail due to harsh conditions and resource limitations. Unlike existing systems, SALINA integrates real-time data analysis with solar-powered operation, enabling continuous wildlife tracking even in challenging weather. For example, SALINA has been used to monitor salmon populations, providing accurate, real-time insights into their behavior, migration patterns, and population dynamics. By enhancing sonar data quality and applying advanced AI models tailored for noisy underwater environments, SALINA addresses critical gaps in tracking species like salmon, which are vital for ecological balance and the cultural and economic well-being of communities. This research is particularly timely as the need for effective, scalable tools to monitor and protect fragile ecosystems is growing amid environmental challenges. By improving our ability to study and manage wildlife, SALINA has the potential to make a lasting impact on conservation efforts and ecological research.

Perspectives

Writing this article has been a rewarding experience, as it combines my passion for technology and conservation. Developing SALINA allowed me to contribute to protecting ecosystems and monitoring critical species like salmon, which play a vital role in both nature and communities. Collaborating with experts from diverse fields has been enriching, and I hope this work inspires further innovation in sustainable wildlife monitoring and conservation.

Chi Xu
Simon Fraser University

This experience has deepened my passion for computer vision, as it allowed us to develop an effective and efficient object detection solution for sonar data. It also provided invaluable hands-on experience with deploying systems on edge devices, addressing real-world challenges like computational performance and resource constraints. I hope this work inspires others to innovate in applying computer vision to complex and noisy data environments, further advancing the field and its applications.

Rongsheng Qian
Simon Fraser University

This has been an incredibly fruitful and rewarding project to work on. The challenge of putting computer-vision AI in the hands of fishery managers and local communities is fundamental to creating a sustainable future for wild salmon and people. To me this work highlights the power of interdisciplinary collaboration, and co-development of research and technical products with Indigenous partner communities. Great work to everyone on the team, and lots more exciting research and on-the-ground outcomes are ahead of us.

William Atlas

Read the Original

This page is a summary of: SALINA: Towards Sustainable Live Sonar Analytics in Wild Ecosystems, November 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3666025.3699323.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page