Line of action:

Status:

Execution date:

General Objective:

Implementation and validation of a recreational fishing monitoring system based on time series of high-definition photographs for the characterization and evaluation of recreational fishing effort in the marine Natura 2000 Network.

Specific objectives:

1. Development and implementation of an autonomous system for the acquisition of high-resolution images for the monitoring of recreational fishing in two areas of the marine Natura 2000 Network.
2. Development of an algorithm for transforming the data generated by the AutoNatura2000 monitoring system into coordinates and of a classifying algorithm of boat typologies (recreational fisherman, cruiser, professional fishing boat, etc.) and boat states (fishing vs. sailing).
3. Validation of the AutoNatura2000 fishing monitoring system: vessel positioning error, percentage of correct classification of vessel typologies and status.
4. AutoNatura2000 Output Generation: maps of the distribution of recreational fishing in the marine Natura 2000 Network, impact habitats, fishing effort and number of recreational fishermen and surveillance of protected areas (% of infringements).
5. Carried out a SWOT analysis for the implementation of the AutoNatura2000 recreational fishing monitoring system in other marine and terrestrial areas of the Natura 2000 Network.

AutoNatura2000 has demonstrated the potential as a vision-based monitoring system for automatic tracking of recreational fishing. The project has installed an autonomous system for acquiring high-resolution images in three areas of the Natura 2000 Network (Marine Protected Area of the Bay of Palma, Isla del Toro and Malgrats Islands). It was concluded that the system can reproduce very faithfully the recreational activity in the Natura 2000 Network, including the detection and positioning of fishing boats, as well as their status (fishing vs. navigation) automatically.

During the execution of the project, a total of 96,364 images were obtained, which were added to the 102,297 of the photographic database of the period 2012-2016 in the Bay of Palma, creating a complete database organized by years, areas and resolution, with a total of 198,661 high-definition photographs.

The next step has been the development of the algorithms of the recreational fishing monitoring system, capable of recognizing, identifying and tracking vessels despite the different weather and light conditions and the number and arrangement of fishing boats. This development has made it possible to locate 93% of the ships. It has also been wanted to discern between boats “fishing” or “sailing” depending on speed, direction and direction. To validate the monitoring system, a test was carried out with a boat equipped with GPS that would allow the data obtained to be compared, the error being not significant.

From the data obtained, practical examples of interest have been generated with the aim of promoting environmentally sustainable fisheries and facilitating the application of the Common Fisheries Policy (CFP) and the Integrated Maritime Policy (IMP), both priorities of the EMFF, thus observing the potential of this monitoring system.

Finally, the internal characteristics (Weaknesses and Strengths) and the external situation (Threats and Opportunities) of the recreational fishing monitoring system developed in AutoNatura2000 have been identified.

In the future, it will be possible to continue working on the improvement of the algorithms generated and on the design of programs, also based on big data technology and deep learning, to automatically determine the number of individuals, identify species, distribute sizes, etc., in an unsupervised and non-invasive way.