AutoFish: AI Phenotyping for Tropical Aquaculture

AutoFish is a JCU-led AI project developing automated phenotyping tools for aquaculture species, with an early focus on barramundi breeding and production systems. The platform combines imaging hardware with lightweight deep learning models to identify body landmarks and estimate traits that are important for grading, selective breeding, and farm decision-making. This allows producers to gather more consistent, objective, and scalable information than manual assessment alone.

The work supports tropical aquaculture by helping farmers and breeding programs identify desirable characteristics linked to growth, quality, and productivity. Because the models are designed for efficient deployment, the technology is well suited to practical use in commercial environments where speed, cost, and reliability matter.

AutoFish sits within JCU’s broader strength in aquaculture innovation and contributes to a future where AI supports more precise, data-driven seafood production. It demonstrates how advanced computer vision can be translated into tangible benefits for northern Australia’s aquaculture industries, including improved breeding workflows, stronger productivity, and better use of digital technologies across the sector.

Project Team and Collaborators:
Professor Mostafa Rahimi Azghadi, Dr Alzayat Saleh, Prof Dean Jerry; The Company One; collaborators linked to JCU aquaculture programs and the ARC Research Hub for Supercharging Tropical Aquaculture through Genetic Solutions

Funding Sources to be Acknowledged:

Australia’s Economic Accelerator (AEA) Ignite Funding