ADVANCE: Automatic Data and Video Acquisition for uNderwater monitoring across Coastal Environments
User group PI: Simone Marini, CNR Istituto di Scienze Marine (CNR-ISMAR), Italy
Hosting infrastructures: SBI Galway Bay Cabled Observatory, Ireland; UPC Expandable Seafloor Observatory, Spain
Many devices exist for the acquisition of underwater images. Most of them have been designed to be towed by supporting vessels or to be installed on cabled observatories. These devices transfer the acquired data through cables and the video/image processing is performed outside the acquisition device (e.g. on board the host vessel or at the laboratory); the power needed for their operation, including the lighting system, is assured by cables. On the contrary, very few devices have been designed to be stand-alone (not wired), autonomous (without human interaction) and suitable for working over extended periods of time. In this context, the aim of this proposal is twofold, technological and ecological at the same time: on one hand we want to test the efficacy of the imaging device GUARD1, described in the European Patent EP2863257 Marini et al. (2013) and in Marini et al. (2015) and Corgnati et al. (2016), as a stand-alone and autonomous sensor capable of quantifying biological activities at individual, population, and community levels. The GUARD1 consists of a low-power system conceived for installation on both fixed and mobile platforms for acquiring images of objects or organisms from 1 mm to 100 cm in size. On-board the device, the image content is autonomously analysed, recognized and classified. Even if the GUARD1 is capable to transmit the information extracted from the acquired images outside the device, in this project, such data will be stored in order to be accessed for further analysis after the recovery. On the second hand we will assess the potentiality of this imaging device by comparing the information automatically produced by the system with the visual inspection of the images acquired by the cameras that the two observatories provide. In particular, the images will be acquired continuously during the day and night (by using the GUARD1 lighting system) for a period of at least two months, in order to estimate the image quality with a different diffusion of light and with different conditions of water turbidity.
The biological assessment of the information produced by the GUARD1 will be obtained through the fulfilment of specific data analysis:
- achievement of faunistic lists based on image recording: this will be mostly focused on fish but also considering other megafaunal species such as echinoderms, cephalopods or large decapods;
- assessment of the minimum number of images required for an efficient sampling of the megafaunal community;
- methodological advances for reliable species identification by still imaging sources in the Mediterranean (OBSEA) and in the Atlantic (CPO) coastal areas; iv) collection of ethological data on species interactions and statistically significant co-occurrences, as a proxy of predator-prey relationships; v) collection of continuous high-frequency visual data for all detected species as proxy of populations’ activity rhythms at both 24-h and seasonal scale; vi) linking biotic data on species assemblages with environmental fluctuations (both oceanographic and meteorological: Fanelli et al. 2013; Fanelli et al. in prep.).
- Corgnati L, Marini S, Mazzei L, et al. Looking inside the Ocean: Toward an Autonomous Imaging System for Monitoring Gelatinous Zooplankton. Martínez J-F, ed. Sensors (Basel, Switzerland). 2016;16(12):2124. doi:10.3390/s16122124.
- Fanelli, E., Cartes, J.E., Papiol, V., López-Pérez, C.. Environmental drivers of megafaunal assemblage composition and biomass distribution over mainland and insular slopes of the Balearic Basin (Western Mediterranean). Deep-sea Res. I 2013, 78: 79-94.
- Fanelli, E., Sbragaglia, V., Azzurro, E., Marini, S., Del Rio, J., Toma, D., Aguzzi J. Drivers of seasonal and interannual changes in coastal fish assemblages by multiparametric observatory monitoring. Mar. Env. Res. Submitted
- Marini, S.; Corgnati, L.; Mazzei, L et al. GUARD1: An autonomous system for gelatinous zooplankton image based recognition. In Proceedings of the OCEANS 2015, Genova, Italy, 18–21 May 2015; pp. 1–7.
- Marini S. et al. Underwater images acquisition and processing system. European Patent EP 2863257A1, 2013, https://data.epo.org/gpi/EP2863257A1-Underwater-images-acquisition-and-processingsystem