Aquaculture meets machine learning. Count sea lice and accurately measure biomass in real-time while reducing cage furniture. Our experts‑in‑the‑loop ensure that every single prediction is correct.SaaS Aquaculture Machine Learning
The last two week-ends I fell into the Aquaculture tech hole (it was totally random), which I had no idea existed. Aquaculture is a huge industry with half of all seafood consumed by human already coming from aquafarms, and according to the FAO, this ratio will reach two-thirds by 2030. A whole aquaculture tech ecosystem is emerging to solve problems around four areas:
Better aquaculture systems where fresh & sea water fishes are farmed.
Fishmeal alternatives to avoid having to feed farmed fish with wild fish…
Biological therapeutics to keep seafood healthy (a huge problem).
Hardware and software to improve operational efficiency: from data analytics solutions to logistics/traceability software.
This week I'm covering Aquabyte a startup using AI to do “fish recognition” in aquafarms a.k.a #surveillance for fish.
The global aquaculture market is already quite big, estimated at $180B in 2018 for more than 75M tons of seafood farmed, and growing. The major production areas are located in Asia (China, Indonesia, Vietnam…) where mostly fresh water fish is farmed in small inland aquafarms, and in Norway where you find huge sea farms producing mainly salmons.
Aquabyte is an aquafarm monitoring solution that uses an underwater camera combined with an AI powered “fish recognition” software to estimate in real time biomass (number of fishes and their health) and detect parasites. The data extracted is used to give insights to the farmer that help him feed / treat the fishes more efficiently.
From what I read, aquafarms (yes, I admit i’m not an expert at farming fish yet) are still mostly run a in a “traditional” way, and this industry is still behind in terms of tech penetration compared to land agriculture. Which means there are a lot of opportunities, but also a lot of go-to-market challenges.
Go-to-market challenges are typical of industries transitioning from a low to a higher level of digitization, startups have to lead market education, convince customers of the value of their solution, and even tougher, change customers habits. This is why timing is key, and that very often the first movers are not the final winners (first movers conduct market education and second movers win with better products).