The proposed pilot study aims to compare two imaging techniques for collecting data on reef fish in the gulf of mexico: high-frequency acoustic imaging sonars and optical stereo-baited underwater…

The proposed pilot study aims to compare two imaging techniques for collecting data on reef fish in the gulf of mexico: high-frequency acoustic imaging sonars and optical stereo-baited underwater video systems (s-bruvs). While traditional optical video surveys are effective, high turbidity in the western gulf often limits their utility. Acoustic imaging, which can capture shape and swimming behavior, is advancing as a viable alternative. By conducting concurrent surveys in high visibility areas of the eastern gulf, the study will generate datasets to evaluate the strengths and limitations of acoustic imaging technology. Advances in deep learning technology will be implemented for automating the post-processing of imaging sonar data and compared to manually annotated optical imagery. This study will also address the common issue of slow processing times for optical methods through exploration of various automated processing routines. Ultimately, this project will aid in determining how data from optical and acoustic imaging sonars can be integrated to generate accurate abundance indices for reef fish, contributing to fisheries-independent data pipelines for stock assessments.