Aquaculture Industry Continues to Adopt Algorithms, Artificial Intelligence

Have you ever thought about what it takes to build an algorithm? An algorithm can be as simple as a set of instructions (such as a recipe) or as complicated as Google’s search engine, which can provide you with a billion relevant web pages in a matter of seconds. Whether you’re baking a cake or asking Google how sound travels underwater, the power of algorithms is incredible – and often misunderstood.

Algorithms are the building blocks of artificial intelligence, which is typically defined as any kind of intelligence demonstrated by machines that aims to mimic human or natural intelligence. Artificial intelligence is an umbrella term that encompasses other data analysis techniques, such as machine learning or statistical modeling, that help us process extremely large amounts of data to derive insights from patterns, trends and relationships between variables.

AI and machine learning are becoming everyday tools in a wide variety of industries, including aquaculture. In the 1980s, the concept of “precision agriculture” was being applied in terrestrial farming with the goal of improving efficiencies in order to feed more people per farm and keep up with growing human populations.

Precision agriculture employed technologies like drones and satellite imagery to ingest new data streams into everyday decision making. The use of sensors to monitor soil moisture and nutrient levels and the use of weather prediction models provided new insights and enabled farmers to make data-driven decisions.

AI in Aquaculture

Similar concepts have been explored in aquaculture. As a relatively new industry, aquaculture has emerged during the era of advanced technology and data analytics and there is an ongoing shift in the industry to adopt smart technology that uses AI and advanced data analytics to predict and provide insights into a wide array of farm operations.

As with its counterpart on land, “precision aquaculture” aims to improve the ability of fish farmers to monitor, control and document key factors affecting fish production. This will allow them to fine tune operational decisions, improve fish health and achieve higher efficiencies and profit.

The goal of AI solutions like those being developed by Innovasea is to provide data-based recommendations for decision making that can be viewed by all stakeholders and provide a rationale for why certain actions should or should not be taken.

Feeding and Beyond

Some of the best applications of this technology center around feeding, which is central to all farm operations.  Feed is the single highest operating cost for fish farms, so small efficiency improvements happening daily can result in significant financial savings over the long term. The concept of data-supported feeding is to consider all parameters that impact not only when to feed, but how much and at what rate.

Special cameras that can analyze hundreds of fish in a few hours to provide highly accurate biomass estimates can help operators better determine how much feed is required in a particular pen. Underwater camera networks combined with species-specific fish satiation algorithms and pellet detection capability can assess feed demand on a second-by-second basis to alert operators when to slow down or stop feeding.

Elsewhere, machine vision algorithms are being used to identify parasites and diseases automatically. That eliminates the need for manual lice counting, which is stressful and inefficient, and provides a more reliable indicator of fish health. These emerging technologies are currently being researched, developed, and validated by companies like Innovasea and many are starting to become part of standard operating procedures at commercial fish farms.

These are just some of the technologies highlighted in an upcoming white paper I co-authored with Innovasea colleagues Rafi Cordero, Raghu Pasula and Srikanth Reddy. The paper, “The Power of Data Science and How It’s Transforming Aquaculture,” examines new AI concepts and technologies that are being developed for fish farms, including new types of imaging systems that can detect sea lice, disease, food pellets and even harmful microscopic organisms like plankton. It also discusses new data that’s available thanks to these emerging technologies and highlights how valuable farm management systems can be if they have access to new insights that would otherwise go undetected.

 

Interested in hearing more? Download the white paper to learn about these exciting developments that are helping fish farmers take better care of their fish stocks and improve production efficiency.