Skip to content
Portada » News » Innovation » The CSIC optimizes advanced techniques for the identification and quantification of fishing catches

The CSIC optimizes advanced techniques for the identification and quantification of fishing catches

Through the IIM (Vigo) and within the SICAPTOR project.

  • Tools based on artificial intelligence have been developed to recognize and quantify species even when they overlap or are superimposed, and autonomous operation has been achieved.
  • These improvements have been validated in more than 80 fishing hauls under real conditions, with promising results and no technical failures.
  • These advances will make it possible, among other aspects, to assess fishing grounds more accurately and improve the efficiency of fleet activity.

Santiago de Compostela, 6 February 2020. CSIC has successfully completed, after just over a year of research, the project “Implementation of an electronic total-catch documentation system for sustainable and online management of fishery resources” (SICAPTOR).

The project was coordinated by the Process Engineering group at the Institute of Marine Research (IIM). Partners included the Spanish Institute of Oceanography (IEO), the Galician Supercomputing Centre (CESGA) and the Fishing Producers’ Organization of the Port and Ría of Marín (OPROMAR). It was supported by Fundación Biodiversidad through the Pleamar Programme, co-financed by the European Maritime and Fisheries Fund (EMFF).

The overall objective was to improve the iObserver system, an electronic device installed above the sorting belt that photographs the entire catch, analyses each image to identify species and estimate the size and weight of each specimen, and sends the information in real time to a land-based server.

According to Luis Taboada, CSIC scientist and principal investigator of SICAPTOR, the challenge was to obtain complete and reliable data on fishing activity, including retained catch and discards. SICAPTOR developed advanced total-catch quantification techniques and implemented artificial-intelligence tools for automatic species recognition.

The main results include autonomous operation of the system, recognition of overlapping specimens, improved estimation of catch volume and the generation of tools that can support sustainable fisheries management.