Abstract
High-quality datasets are crucial for computer vision in endangered fish monitoring. Species similarities and dynamic environments pose challenges. The Pomeranian region, a key salmonid refuge with ongoing restoration efforts, necessitates robust monitoring to assess restoration success. To facilitate environmentally sustainable research, we introduce a dataset, Pomerania Fish (PomerFish), comprising two sub-datasets: PomerFishObj and PomerFishSeg, each with corresponding annotations. This dataset, compiled by salmonid experts, focuses on endangered Central European salmon populations. It was collected from 2015 to 2024 in the Pomerania Region, Central Europe, using a GoPro Hero 5 camera. The dataset comprises: (1) PomerFishObj, containing 14,989 high-resolution images with manually annotated bounding boxes and 3,273 negative samples (fish absence); and (2) PomerFishSeg, containing 1,115 images, including 1,038 with polygon-based segmentation masks and 77 negative samples. Characterized by high-resolution imagery, large data volume, and comprehensive habitat records beyond species, it enables training for detailed underwater observations and precise species growth estimations. This dataset supports targeted conservation and habitat management, providing crucial resources for research, species detection, and conservation practices.
Data availability
The code, model, and associated notebooks for data processing and baseline computation are available in our Zenodo repository: https://zenodo.org/records/17432128.42. The software toolkit, developed using Python 3.10 and the PyTorch 2.3 deep learning framework, offers the following functionalities, including data preprocessing, model training, model evaluation, deployment and inference, and visualization.
References
Tye, S. P., Fey, S. B., Gibert, J. P. & Siepielski, A. M. Predator mass mortality events restructure food webs through trophic decoupling. Nature 626, 335–340 (2024).
O’Brien, A., Townsend, K., Hale, R., Sharley, D. & Pettigrove, V. How is ecosystem health defined and measured? A critical review of freshwater and estuarine studies. Ecological Indicators 69, 722–729 (2016).
Thomas, P. O. et al. Electrofishing as a potential threat to freshwater cetaceans. Endang Species Res 39, 207–220 (2019).
Curry, R. A., Hughes, R. M., McMaster, M. E. & Zafft, D. J. Coldwater fish in rivers. Standard methods for sampling North American freshwater fishes. American Fisheries Society, Bethesda, Maryland 139–158 (2009).
Raby, G. D., Colotelo, A. H., Blouin-Demers, G. & Cooke, S. J. Freshwater Commercial Bycatch: An Understated Conservation Problem. BioScience 61, 271–280 (2011).
Bielli, A. et al. An illuminating idea to reduce bycatch in the Peruvian small-scale gillnet fishery. Biological Conservation 241, 108277 (2020).
Hinlo, R., Lintermans, M., Gleeson, D., Broadhurst, B. & Furlan, E. Performance of eDNA assays to detect and quantify an elusive benthic fish in upland streams. Biological Invasions 20, 3079–3093 (2018).
Jothinarayanan, N., Pham, C. H., Karlsen, F. & Roseng, L. E. eDNA-Based Survey of Fish Species in Water Bodies Using Loop-Mediated Isothermal Amplification (LAMP) for Application of Developing Automatic Sampler. Methods and Protocols 7 (2024).
Ilarri, M., Souza, A. T., Dias, E. & Antunes, C. Influence of climate change and extreme weather events on an estuarine fish community. Science of The Total Environment 827, 154190 (2022).
Le Hen, G. et al. Alien species and climate change drive shifts in a riverine fish community and trait compositions over 35 years. Science of The Total Environment 867, 161486 (2023).
Parmesan, C. & Yohe, G. A globally coherent fingerprint of climate change impacts across natural systems. Nature 421, 37–42 (2003).
Prakash, S. Impact of Climate change on Aquatic Ecosystem and its Biodiversity: An overview. International Journal of Biological Innovations 3 (2021).
Woodward, G., Perkins, D. M. & Brown, L. E. Climate change and freshwater ecosystems: impacts across multiple levels of organization. Philos Trans R Soc Lond B Biol Sci 365, 2093–2106 (2010).
Xu, L. et al. Asymmetric impacts of climate change on thermal habitat suitability for inland lake fishes. Nature Communications 15, 10273 (2024).
Czerniawski, R. & Bilski, P. Funkcjonowanie i Ochrona Wód Płynących 2023 (2023).
Piskuła, P. & Astel, A. M. Microplastics Occurrence in Two Mountainous Rivers in the Lowland Area—A Case Study of the Central Pomeranian Region, Poland. Microplastics 1, 167–185 (2022).
Roy, A. M., Bhaduri, J., Kumar, T. & Raj, K. WilDect-YOLO: An efficient and robust computer vision-based accurate object localization model for automated endangered wildlife detection. Ecological Informatics 75, 101919 (2023).
Tanwari, K., Terefenko, P., Śledziowski, J. & Giza, A. Manually Annotated Drone Imagery Dataset for Automatic Coastline Delineation. Scientific Data 12, 383 (2025).
Ditria, E. M., Sievers, M., Lopez-Marcano, S., Jinks, E. L. & Connolly, R. M. Deep learning for automated analysis of fish abundance: the benefits of training across multiple habitats. Environ Monit Assess 192, 698 (2020).
Schneider, S. & Zhuang, A. Counting Fish and Dolphins in Sonar Images Using Deep Learning, https://doi.org/10.48550/arXiv.2007.12808 (2020).
Saleh, A., Sheaves, M., Jerry, D. & Rahimi Azghadi, M. Applications of deep learning in fish habitat monitoring: A tutorial and survey. Expert Systems with Applications 238, 121841 (2024).
R. Mandal, R. M. Connolly, T. A. Schlacher, & B. Stantic. Assessing fish abundance from underwater video using deep neural networks. in 2018 International Joint Conference on Neural Networks (IJCNN) 1–6, https://doi.org/10.1109/IJCNN.2018.8489482 (2018).
Kandimalla, V. et al. Automated Detection, Classification and Counting of Fish in Fish Passages With Deep Learning. Frontiers in Marine Science 8 (2022).
Zhu, X., Vondrick, C., Fowlkes, C. C. & Ramanan, D. Do We Need More Training Data? International Journal of Computer Vision 119, 76–92 (2016).
Lévêque, C., Oberdorff, T., Paugy, D., Stiassny, M. L. J. & Tedesco, P. A. Global diversity of fish (Pisces) in freshwater. in Freshwater Animal Diversity Assessment (eds Balian, E. V., Lévêque, C., Segers, H. & Martens, K.) 545–567, https://doi.org/10.1007/978-1-4020-8259-7_53 (Springer Netherlands, Dordrecht, 2008).
Prodhan, S., Diip, N., Akter, S., Farhaan, S. & Mansoor, N. Advancing Fish Species Identification in Bangladesh: Deep Learning Approaches for Accurate Freshwater Fish Recognition. in 113–122, https://doi.org/10.1007/978-981-99-8349-0_10 (2024).
Shah, S. Z. H. et al. Fish-Pak: Fish species dataset from Pakistan for visual features based classification. Data in Brief 27, 104565 (2019).
Ataullha, M., Jisun, S. P., Jahan, I. & Rahman, M. BanFish: A Dataset for Classifying Common Bangladeshi Fish Species. 2024 6th International Conference on Electrical Engineering and Information & Communication Technology (ICEEICT) 335–340 (2024).
Hitt, N. P., Kessler, K. G. & Letcher, B. Annotated fish imagery data for individual and species recognition with deep learning: U.S. Geological Survey data release, https://doi.org/10.5066/P9NMVL2Q (2021).
Majumder, A., Rajbongshi, A., Rahman, M. & Biswas, A. A. Local Freshwater Fish Recognition Using Different CNN Architectures with Transfer Learning. International Journal on Advanced Science Engineering and Information Technology 11, 1078–1083 (2021).
Deka, J., Laskar, S. & Baklial, B. Automated Freshwater Fish Species Classification using Deep CNN. Journal of The Institution of Engineers (India): Series B 104, 603–621 (2023).
Christin, S., Hervet, É. & Lecomte, N. Applications for deep learning in ecology. Methods in Ecology and Evolution 10, 1632–1644 (2019).
Feng, J. & Li, J. An Adaptive Embedding Network with Spatial Constraints for the Use of Few-Shot Learning in Endangered-Animal Detection. ISPRS International Journal of Geo-Information 11 (2022).
VideoLan. VLC media player. https://www.videolan.org/vlc/index.html (2006).
Sekachev, B. et al. opencv/cvat: v1.1.0. Zenodo https://doi.org/10.5281/zenodo.4009388 (2020).
Paszke, A. et al. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems 32 (2019).
Redmon, J., Divvala, S., Girshick, R. & Farhadi, A. You Only Look Once: Unified, Real-Time Object Detection. in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 779–788, https://doi.org/10.1109/CVPR.2016.91 (2016).
Khanam, R. & Hussain, M. What is YOLOv5: A deep look into the internal features of the popular object detector. ArXiv abs/2407.20892 (2024).
Shelhamer, E., Long, J. & Darrell, T. Fully Convolutional Networks for Semantic Segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 39, 640–651 (2017).
Heryadi, Y. et al. The Effect of Resnet Model as Feature Extractor Network to Performance of DeepLabV3 Model for Semantic Satellite Image Segmentation. in 2020 IEEE Asia-Pacific Conference on Geoscience, Electronics and Remote Sensing Technology (AGERS) 74–77, https://doi.org/10.1109/AGERS51788.2020.9452768 (2020).
Chen, L.-C., Papandreou, G., Schroff, F. & Adam, H. Rethinking Atrous Convolution for Semantic Image Segmentation, https://doi.org/10.48550/arXiv.1706.05587 (2017).
Shi, X. et al. PomerFish: A dataset for fishes across Pomerania freshwater waterbodies in-situ environments. Zenodo https://doi.org/10.5281/zenodo.17432128 (2025).
Acknowledgements
This research was co-financed by the Minister of Science under the “Regional Excellence Initiative” Program for 2024–2027 (RID/SP/0045/2024/01). We extend our sincere gratitude to Zhe Shi of the School of Electronic Information, Xi’an Polytechnic University, for her invaluable assistance in dataset development.
Author information
Authors and Affiliations
Contributions
X.S. conceived the research and wrote the original draft. A.F. collected the original data. X.S., T.K. and R.C. reviewed the original data. X.S. developed the dataset. K.T. provides technical support for the dataset and assists with the visualization. R.C. and T.K. supervised the dataset. R.C. provided financial support for this work. All authors reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interest
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Reprints and permissions
About this article
Cite this article
Shi, X., Czerniawski, R., Tanwari, K. et al. Pomerania Fish: A dataset for fishes across Pomerania freshwater waterbodies in-situ environments.
Sci Data (2025). https://doi.org/10.1038/s41597-025-06393-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41597-025-06393-8
Source: Ecology - nature.com
