AI for aquaculture
Image- and video-based methods for objective, scalable monitoring of fish health, welfare, and behaviour.
Computer vision · Artificial intelligence · Aquaculture
Researching how machines can see beneath the surface.
I am a Senior Research Scientist at SINTEF Ocean and Adjunct Associate Professor at NTNU. My work combines computer vision, machine learning, and biological insight to develop non-invasive methods for monitoring salmon welfare.
Research
Vision systems for complex, real-world environments
My research spans mathematical foundations and applied computer vision. A current focus is extracting reliable information about individual fish from challenging underwater imagery—without disturbing the animals being observed.
Biometric re-identification
Recognising individual salmon from natural visual characteristics to support long-term welfare histories.
Motion and 3D analysis
Optical flow, tracking, stereo vision, and reconstruction for understanding movement, shape, and behaviour.
Pattern recognition
Statistical object recognition, graph matching, subgraph matching, and robust visual representations.
Mathematical optimisation
Convex programming and relaxation methods for difficult correspondence, grouping, and image-analysis problems.
Student research
Co-supervision of applied AI and computer-vision projects that connect methodological research with aquaculture needs.
Featured project · NFR 344022
cAIge
Computer Vision and Artificial Intelligence based Salmon Identification and automated long-term welfare assessment in aquaculture net-pens.
The project develops technology for non-invasive identification of individual salmon and continuous, objective welfare monitoring in full-scale aquaculture.
Explore cAIge and its student projects
2023–2027
Research Council of Norway project, led by Christian Schellewald
Selected work
Recent publications
A small selection is shown here. See Google Scholar for the complete and continuously updated publication record.
Patch ensembles for robust salmon re-identification with weak trajectory labels ↗
E. U. Høgstedt, C. Schellewald, A. Stahl, and R. Mester · Accepted for IEEE ICIP, 2026 · arXiv:2605.18038
A multi-purpose tracking framework for salmon welfare monitoring in challenging environments ↗
E. U. Høgstedt, C. Schellewald, A. Stahl, and R. Mester · ICCV Workshops, 2025
Automated computer vision based individual salmon (Salmo salar) breathing rate estimation (SaBRE) for improved state observability ↗
E. B. Høgstedt, C. Schellewald, R. Mester, and A. Stahl · Aquaculture, 2025
Image-based quantification of scale loss in fish using machine learning and computer vision ↗
T. Fjorden, E. B. Høgstedt, C. Schellewald, R. Mester, M. Remen, A. V. Nytrø, and A. Stahl · ICMV 2024 proceedings, published 2025
Robust hole-detection in triangular meshes irrespective of the presence of singular vertices ↗
M. Yip, A. Stahl, and C. Schellewald · Computer-Aided Design, 2024
Mouth opening frequency of salmon from underwater video exploiting computer vision ↗
C. Schellewald, A. Saad, and A. Stahl · IFAC-PapersOnLine, 2024
Active pose — autonomous hole-filling techniques for 3D surface reconstruction in underwater exploration ↗
M. Yip, C. Schellewald, T. Gambin, and A. Stahl · IFAC-PapersOnLine, 2024
Approaches enabling underwater autonomy and sensing in sea-based aquaculture settings ↗
M. B. Skaldebø, C. Schellewald, L. D. Evjemo, H. B. Amundsen, M. Xanthidis, and E. Kelasidi · MED, 2024
Teaching & supervision
Connecting methods with meaningful applications
At NTNU's Department of Engineering Cybernetics, I teach visual analytics and automation and co-supervise master's and PhD candidates working in computer vision, AI, aquaculture, and automation.
The cAIge project provides a strong framework for student research, connecting methodological questions with data and challenges from real aquaculture settings.
TTK37 Visual Analytics and Automation →