Computer vision for aquaculture
TilapAI reads Average Body Weight (ABW) straight from underwater video through our stereoscopic cameras and proprietary computer vision algorithms.
The status quo
To track growth, farmers need to net thousands of fish out of the water with an average cadence of every two weeks, weigh them, and throw them back after counting. This allows the ABW to be calculated. It stresses the fish, drives up mortality and the chance for disease, and only samples a small fraction of the cage.
Weight distribution
Illustrative cage: each bar counts fish in a 5 g band.
Introducing Niloscale
A stereo camera records the cage underwater for two hours. Computer vision picks out every fish it can measure, the camera's depth gives its true length, and a length-weight relationship turns that into a weight.
Fish are never handled, so the measurement itself no longer drives mortality.
A bulk weigh-in divides one total weight by one count. Niloscale weighs each measured fish, so you see how sizes spread across the cage.
Accurate ABW tightens the feed conversion ratio.
Your footage, its processing and your dashboard all run on European providers, by choice.
How it works
A resilient, EU technology stack designed for remote farms. Everything in the field is kept simple and robust, so it keeps running in places where the nearest technician can be days away.
A portable stereoscopic camera unit records each cage over a two-hour window before moving on.
Collected footage is uploaded to EU storage over an encrypted connection.
Proprietary algorithms run across all the collected footage in the EU cloud, measuring every suitable fish and calculating ABW for the cage.
A clean dashboard puts current and historical ABW and weight distributions in the farmer's hands, easy to look back on and use to fine-tune feeding.
Where we're at
Niloscale is moving out of the workshop and into the water. The camera rig is built, the detection model is trained and the dashboard is live. Next, the first rig goes into a commercial cage.
Secured Partners for Water funding to build Niloscale.
Field prototype built and the first detection model trained.
First footage from a commercial farm on Lake Victoria.
New stereo camera, length-based weighing and a live farmer dashboard.
Extended field trials across farms in Uganda, Kenya and Zambia.
Founders
Founder
Floris is the computer science half of TilapAI. He has spent nearly a decade as a software engineer at big tech companies, specialising in cryptography.
He holds a BSc and MSc in finance and an MSc in Computer Science from the University of Bath, with a thesis on detecting disease in tilapia using computer vision.
Co-founder
Sam is the hands-on half of TilapAI. Handy with tools, he has worked on Niloscale's hardware since day one, turning prototypes into kit that holds up on a working fish farm.
Proudly supported by
Niloscale is funded through Partners for Water, a Dutch government programme enhancing water security internationally, run by the Netherlands Enterprise Agency (RVO).
About the programme
Get in touch
We'd love to hear from you.
Amsterdam, The Netherlands