AI4FISH - AI for sustainability in marine recreational fishing participation: Scalable spatio-temporally resolved ML models

We have realized the progress in achieving the project primary objective, have realized SO 1 and partly SO 3, have reached the milestones M1, M2, partly M4. ML models generated map of fishing spots for a part of Norwegian coast have been produced. Algorithms and software pipeline with functionality from data downloading large geospatial dataset until production of maps of fishing spots have been created. With primary datasets, algorithms, and software in place, we do not see unexpected risk factor, although certainly we strive to continuously expand datasets, improve ML models and optimize code. A conference paper abstract entitled Geometric 4D deep learning for fish behavioral intelligence have been submitted but full paper submission can only be expected Q1 2022. Bigdata and large-scale ML platforms such as Hadoop and Apache Spark using either local computer clusters or cloud resources have been evaluated for SOs 4 and 5. Technical details are briefly provided as below. A train and test dataset for part of Norwegian coastal region was integrated. Raw data in different geospatial data formats and different spatial resolution was transformed to raster data, interpolated to a spatial resolution of 100 m, and features were computed. ML models were trained in supervised learning paradigm using manual labels as ground truth and evaluation metrics F1 scores and accuracy above 0.95 were obtained on test data. Generalization of the ML models were evaluated on new unseen data, a lower F1 score of 0.67 and high accuracy of 0.98 was obtained. The prediction results projected to their geospatial coordinates were written to geodata file which can be published as maps on geographic information system software such as Qgis. Techniques combining deep generative models combined with inversion procedure and traditional models were considered to tackle missing data and lack of data in order to cover the whole coastal region.

Recreational fishing (RF) involves at least 226 million participants worldwide, it generates an annual economic value of 5.9 billion euros in EU and 29 billion euros in the US. Until recent years, studies show that RF has nearly invisibly contributed to approximately 12% of global fishing removal; and it has caused 27% to declining in stock of two endangered species. We have identified 5 main barriers in achieving sustainability in RF: - No or limited real-world georeferenced and time-tagged catch data. - No or limited tools for monitoring and control RF activities. - No or limited communication channels. - Limited knowledge about where to fish. - Post-reactive, not predictive or proactive. The FiskHer App could be an good tool in tackling the 5 barriers towards sustainable RF. We develop a new digital service – FiskHer.ai – leveraging state-of-the-art data science and machine learning technologies for promoting and facilitating of sustainable marine recreational fishing. The idea is to discover scalable machine learned models that accurately predict most probable RF spots with high spatial and temporal resolution in Norway and worldwide. We achieve this via synergized transformation of the domain expertise earned in field owned by FiskHer AS to machine learned models, which are incrementally learned continuously lifelong with user contributed new data. The expected results of the project are a new integrated large dataset with validated labels, new underwater 360 degree video and hyperspectral imaging data for a few fishing spots, software for data integration and automated machine learning, scalable machine learned models for accurate and spatio-temporal resolved predictions. This innovation is most important to us because: 1. it allows us to go from static to dynamic areas. 2. It gives us a wonderful tool that will save us years of manual registration. It will thus be cost-effective and innovative for us, and enable us to deliver many years ahead of schedule.