Machine learning and vision startup Aquabyte announced Tuesday that it has raised $3.5 million in seed funding in order to build out a team of developers to refine its software, which is geared toward using machine learning and vision to reduce costs for fish farming.
The round was co-led by Costanoa Ventures and New Enterprise Associates. Princeton University and other investors also participated in the round.
“The development of computer vision over the past couple years along with the advent of deep learning has opened up dramatic opportunities to build new vision-related products that can solve very practical, real world problems,” said Bryton Shang, Founder and CEO of Aquabyte. “The same computer-vision models I worked on for tissue cancer diagnosis are applicable in a way that can transform the fish farming industry and the future of protein consumption around the world.”
The main purpose of the round is to help the company build out a team to develop its technology, both at its headquarters in San Francisco as well as Norway, where the company is working with pilot customers. The company is also focused on Norway because there’s significantly more fish farming there than there is in the United States.
Initially, the company is focused on developing its machine learning and machine vision software to develop two algorithms: one to determine the size of salmon over time, and the other to determine the presence of sea lice, a parasite sometimes found on salmon. But it won’t be easy.
That rich problem is something the company is working to solve in a number of areas, both in fish farms in net pens in open water as well as tanks – a growing area of aquaculture. The data is gathered using underwater 3D cameras.
“We’re in the middle of building those algorithms,” Shang said. “Then we’ll commercialize those and use the data to develop feeding algorithms.”
The company’s ultimate goal is to use those feeding algorithms to more finely control the amount of food that fish farmers use. If it’s successful, the company claims it could save fish farmers 20-30% of the cost of food currently used. That’s a big deal, considering that feed accounts for about half the cost of the average fish farm.
Once the company has optimized its algorithms for feeding salmon, it intends to move on to other kinds of fish as well as other markets.
“Our plan is to start in the Norwegian market then expand to Chile, Canada and Scotland,” the company wrote in a briefing document. “The same technology is applicable to other species such as trout, seabass, and seabream.”