Using AI-based image analysis to quantify and optimise the hatching rate of Artemia salina

Abstract

This thesis investigates how varying levels of salinity and the application of dechlorinating agents affect the hatching success of Artemia salina, a high-value commodity in aquaculture. It uses an AI-driven RF-DETR detection methodology and identifies a physiological hatching optimum at a salinity of 19.21 ppt, where a peak hatch rate of 80.2% was achieved. At this concentration, a near-isotonic state was reached, minimising the metabolic energy required for glycerol accumulation. The results demonstrate that while the application of dechlorinating agents enhances hatching success at lower salinities (17.59 ppt) by reducing oxidative stress and glycerol requirements, this benefit diminishes as osmotic pressure increases. Ultimately, these findings suggest that salt concentration optimisation is more critical for maximising Artemia hatching than chemical conditioning alone.