Системы подводной видеосъемки для изучения численности и разнообразия зоопланктона: проблемы и возможности

Underwater zooplankton video imaging systems have been reviewed in the context of their development, methods, and applications, with a more detailed focus on systems that have appeared in the last five years. Three notable trends are observed: 1) reduction in the size and price of systems; 2) transition from expensive equipment and licensed software to free hardware and software; 3) expanding the capabilities of systems to be applicable both in the laboratory and in the field. Extraction of ecological and biological information from underwater videos is greatly facilitated by a variety of machine learning techniques; however, their accuracy is highly dependent on the quality of the training datasets. A new optic-fluorescent flow-through system for zooplankton assessment (ZooFluoBox) is presented to expand the range of the parameters measured by underwater imaging instruments and enhance the control of turbulent and light conditions during video recording. Fluorescence detection allows distinguishing between eating and non-eating zooplankters: diapausing or dead and live ones. The system is designed to profile zooplankton abundance and depth distribution, mostly in inland waters, as its depth limit is 60 m and object size limit is 5 mm