Datasets for Aquaculture: A Deep Learning Perspective

Aquaculture is being used to a greater extent for conservation goals, such as species recovery, habitat restoration, and mitigating the effects of wild capture on harvested species that are vulnerable. Deep Learning (DL) models are in use for training datasets of aquatic organisms, to provide solutions to various problems in aquaculture farming. However, there is still motivation, awareness, and knowledge that are found necessary to encourage researchers to explore the potential unmet. In this work, the various aquaculture-associated datasets used in DL for image processing procedures have been elucidated. It addresses the challenges of these datasets to pinpoint areas for improvement, ultimately paving the way for future advancements in the application of DL within aquaculture.