Towards a Fully Automated Underwater Census for Fish Assemblages in the Mediterranean Sea
- Others:
- Ecology and Conservation Science for Sustainable Seas (ECOSEAS) ; Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)
- Modèles et algorithmes pour l'intelligence artificielle (MAASAI) ; Inria Sophia Antipolis - Méditerranée (CRISAM) ; Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Laboratoire Jean Alexandre Dieudonné (LJAD) ; Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Scalable and Pervasive softwARe and Knowledge Systems (Laboratoire I3S - SPARKS) ; Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis (I3S) ; Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis (I3S) ; Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Centre National de la Recherche Scientifique (CNRS)
Description
Assessing underwater biodiversity is a labour-intensive and costly procedure whilst being crucial to measure the extent of local fish stock declines. In most cases, Underwater Visual Census (UVC) is the method of preference, however this can be human-costly and is limited by meteorological and logistic factors. Advances in technology allows the utilisation of more autonomous video recording methods (i.e. Remote Operated Vehicles (ROV)) which work around the aforementioned limitations. This study used a transect-wise UVC coupled with diver operated videos (DOV) simulating an ROV. For the video analysis, a comprehensive fully automated pipeline was developed to extract frames from DOV and perform color correction. This pipeline integrates a YOLO-based model for the detection of 20 Mediterranean fish species validating presence or absence of each species within individual transect. This study was conducted to evaluate the feasibility of utilising video-based methods for UVC with minimal human-dependence. The automation of the video analysis showed accordance with the manual video counting enabling an autonomous and bias-free procedure for video assessment. In conclusion, utilising a minimal-human-dependent video method disconnects the data acquisition from limiting factor (i.e. meteorological and logistic) and automation of this video analysis will significantly reduce the labour and time required by researchers. For future 1 fieldwork campaigns, the video data collection protocol needs to be adjusted to better resemble the traditional UVC and bring forward this acquisition method.
Additional details
- URL
- https://hal.science/hal-04690514
- URN
- urn:oai:HAL:hal-04690514v1
- Origin repository
- UNICA