About
SeaBee Infrastructure
SeaBee is structured to secure an efficient establishment, running and operation of the infrastructure during the first five years (Establishment Phase), the following five years (Operational Phase) and beyond. SeaBee is funded by the Norwegian Research Council (No. 296478 in Prosjektbanken)
We specify user requirements, design drone and sensor infrastructure, purchase hardware and software, and set up physical space.
Lead: Anders Gjørwad Hagen (NIVA)
Two physical nodes are planned at NIVA in Oslo and NTNU in Trondheim, where the required space is guaranteed by the two institutions.
SeaBee identifies user requirements, including specifications to purchase drones and sensor systems, and allocates the space necessary to accommodate drones, sensors, and associated infrastructure.
Systems proven in the field are preferred. Fixed-wing and multirotor drones are required for different applications, e.g., fixed-wings cover larger areas in shorter time (300 h/hour), while rotor drones are preferred for low altitude (<50m) and high precision mapping (<1cm/pixel).
We define, specify and purchase drone-sensor systems that fit the needs of downstream infrastructure. We systematically evaluate requirements based on feedback from the other infrastructure components
We facilitate training of drone pilots, secure registration practices and legislation. We produce safe operation protocols and emergency manual procedures.
Visit the Tech page to see the available drones and sensors.
Lead: Tor Arne Johansen (NTNU)
We ensure the SeaBee drone fleet is operational and efficient at all times by training drone pilots, integrating sensors on drones, calibrating and testing equipment and educating personnel for maintenance and repair.
We register RO 1, 2, and 3 pilots for VLOS (Visual Line Of Sight), EVLOS (Extended Visual Line Of Sight), and BLOS (Beyond Line Of Sight) flights.
Sensors are integrated, calibrated, and tested. High quality data are validated by state-of-the-art calibration bench ESA’s Fiducial Reference Measurements for Satellite Ocean Color (FRM4SOC), owned by NIVA.
Technical workshops for drone maintenance and repair maintain and repair the drones where they are stored to reduce operational expenses and minimize operational risks. Documentation on repair procedures is supplied.
- State-of-the-art sensors for radiance measures and imaging collection, including visual-light RGB cameras, and sensors for multispectral and hyperspectral imaging, thermal, near-infrared-red, and solar-induced fluorescence.
- Flexible solutions adapted to the rapid developing technology in the field. For example, accurate navigation (e.g. RTK GPS), communication hardware for on board logging of image and metadata, and real-time viewing of image acquisition by ground control stations and drone operators.
- High-performance computing hardware (CPU/GPU) for image pre- and post-processing including geo-referencing, automated quality control and data product generation rented from the national e-infrastructure UNINETT/Sigma2.
- Custom solution software to handle the above in the processing pipeline will be installed in partner computer facilities.
- Hardware platforms and cloud-based interface facilities for web-based data handling and sharing at UNINETT/Sigma2.
We perform image analysis and implement a data workflow structure by building pipelines, as well as evaluating the data analysis.
We accommodate automated image analysis in a plug-and-play manner via advanced and cutting-edge Artificial Intelligence (AI) algorithms.
Lead: Arnt-Børre Salberg (NR)
Data analysis is the core of the SeaBee product generation engine.
We improve and develop within image and object analysis for identification of environmental and pollution relevant variables to respond to new technologies.
Drone data is uploaded to the SeaBee data infrastructure, pre-processed and prepared for storage and analysis. The pre-processing approach contains:
- orthorectification
- image stitching
- radiometric calibration quality assessment of the data
- removal of personal/sensitive information and metadata
Products are generated by state-of-the-art machine learning (ML) and other artificial intelligence (AI) protocols for object detection and thematic mapping of environmental data and creation of visualization products. For example, marine habitat maps in coastal bays and inlets and automated recognition of marine mammals.
Successful implementation of state-of-the-art ML algorithms is dependent on graphics processing units (GPUs) and fast-working storage provided by the UNINETT/Sigma2 high performance computation infrastructure.
A key part of the infrastructure is a framework that enables users to teach an algorithm to perform a specific task, to create a desired product, or identify objects of interest.
We design, establish and implement cloud-based data storage solutions and cutting-edge web-based data sharing applications running on GIS-based visualisation tools.
Lead: James Edward Sample (NIVA)
We are responsible for up-to date data management and ensuring that SeaBee responds to national regulations and recommendations on data policy (RCN 2017).
SeaBee will develop and establish a cloud database for storage of raw, processed and metadata using appropriate and scalable database solutions to ensure fast and efficient data access.
A “cloud agnostic” solution based on existing technologies, with UNINETT/Sigma2 as the hosting provider, will ensure a cost-efficient and validated technical solution for safe and long-term data storage, sufficient to manage large amounts of data, as envisioned being produced by the infrastructure.
Approximately 50 to 100 TB of data is expected within year 1 to 5, and up to 500 TB after 10 years.
Using novel Application Programming Interface (API) technology, data will be shared openly and free of charge. API technology facilitate integration with third party solutions. A user-friendly front-end portal allows interactive search, discovery and downloading functions of SeaBee cloud-based data.
We collect data on the ground using traditional benchmark methods that validate drone products and guide and improve automated product generation.
Lead: Kristina Kvile (NIVA)
It is essential during the infrastructure Establishment Phase to ensure high quality imaging products. As we approach the Operational Phase, the process will become more automated and data collection on the ground will become less central for the included applications.
We provide product validation through data ground-truthing and develop protocols for drone-based automatic procedures to identify coastal habitats, seabird and marine mammal communities, water quality, and anthropogenic impacts. We will carry out several field campaigns to collect targeted ground-truth data in the following topics:
Coastal habitat mapping (led by NIVA)
Marine Mammals (led by NTNU)
Seabirds (led by NINA)
Water Quality (led by NIVA)
The data product validation takes advantage of existing national data collection programs with external funding to support the field campaigns, for example, ØKOKYST (monitoring coastal ecosystems), SEAPOP (mapping and monitoring seabirds), NorSOOP (surface water applications), MERCES (marine restoration).
We disseminate SeaBee results to selected target audiences through various channels.
Lead: Lorna Little (GRID-Arendal)
Communication, outreach and branding involves efficient end-user communication, raising awareness of SeaBee and cultivation of working relationships with stakeholders.
We produce a variety of communication products, like videos, branding material, news items and more.
The audiences we aim to reach include educational and research institutions, governments, consulting companies, industries, NGOs and the general public.
The stakeholders are reached through tailored channels, such as the website, social media, workshops, and written briefs.
Targeted communication strategies and dissemination will allow for comprehensive impact analyses.
We are responsible for the SeaBee infrastructure coordination and risk management.
Lead: Kasper Hancke (NIVA)
Co-lead: Hege Gundersen (NIVA)
We work to coordinate all partners to deliver different parts of the SeaBee service and infrastructure.
We communicate the vision and goals of the SeaBee service and infrastructure.