JERUSALEM, Jan. 12 (Xinhua) -- Israeli researchers said on Monday they have developed an artificial intelligence (AI) system that detects fish hunger by analyzing the sounds they make while eating, offering a new tool for smarter and more sustainable aquaculture.
The team from the National Center for Mariculture (NCM) in Eilat and the Israel Institute of Technology trained machine learning algorithms to identify the distinct sound sea bream make when biting food pellets, filtering out background farm noise. This allows the system to monitor feeding activity in real time and assess whether fish are hungry, full, or stressed.
Feeding control is crucial in aquaculture, the researchers said. Yet currently, fish farms usually rely on general estimates based on weight and feeding charts, which can lead to overfeeding -- increasing costs and polluting water -- or underfeeding, which stresses fish, slows growth, and raises disease risk.
The sound-based method, to be published in the upcoming March issue of Computers and Electronics in Agriculture, provides a scalable, automated and cost-effective way to optimize feeding and monitor fish welfare.
The approach has also shown promise in detect stress and aggression in shrimp and oysters, opening the door for wider application in aquaculture management. ■
