Machine Vision Applications for Welfare Monitoring in Aquaculture: Challenges and Opportunities

Machine vision technologies are increasingly being applied in aquaculture to enhance animal welfare monitoring through automated, non-invasive, and continuous observation systems. This article explores the current state and future potential of machine vision in detecting key welfare indicators such as fish behaviour, health status, growth rates, and environmental conditions. It highlights the integration of computer vision with artificial intelligence to enable real-time decision-making and reduce human error. Despite promising advancements, the article identifies several challenges, including data quality, underwater imaging limitations, species-specific variability, and the need for robust validation frameworks. The authors emphasise the importance of interdisciplinary collaboration to overcome these barriers and unlock the full potential of machine vision for sustainable and welfare-oriented aquaculture systems.