Challenge
In many complex production and rearing processes, full automation with classical technical solutions is not feasible. In aquaculture in particular, operational management requires deep expert knowledge:
- Biomass estimation is typically done manually and is labour-intensive
- Manual sampling causes stress to animals and carries contamination risks
- Without real-time data, feed management and facility control remain reactive
Our Approach
SWMS Consulting combined computer vision, machine learning and IoT sensor technology to create a non-invasive, automated system for continuous biomass monitoring. Cameras capture the animals in the tank; AI models count individuals, estimate sizes and derive total biomass in real time.
What We Delivered
- Computer vision pipeline for individual detection and size measurement
- Machine learning models trained on aquaculture-specific image data
- IoT edge device integration for local inference and data transmission
- Dashboard for real-time biomass monitoring and trend analysis
- Integration with facility control systems for automated feed management
Result
Manual sampling effort was dramatically reduced, feed efficiency improved through data-driven control, and the operator gained continuous, objective insight into the biological state of the facility – enabling more sustainable and profitable aquaculture operations.