This paper addresses the challenges of underwater Visual Odometry (VO) by introducing a novel dataset collected with a monocular camera in a controlled pool environment. The dataset is complemented by preprocessing techniques that mitigate underwater imaging issues such as color distortion and blue-green color cast. A comparative analysis of feature detectors identifies AKAZE as the most effective for underwater
imagery. Additionally, a genetic algorithm (GA) optimizes the RANSAC inlier threshold, enhancing essential matrix estimation. Experimental results highlight a significant reduction in root mean square error (RMSE), with AKAZE achieving a 0.07-meter error compared to traditional methods. This study provides advancements in underwater navigation and mapping, promising improvements for autonomous underwater vehicles (AUVs) and marine research., This work was supported by funding from multiple sources: AF received a
Pre-doctoral fellowship from AGAUR (ref. BDNS 474817) and additional
support from the European Commission’s HORIZON-INFRA-2021-SERV-
01 program through the iMagine project (grant agreement 101058625). The
research also utilized the EGI infrastructure with dedicated support from EGICESGA-
STAC., Peer Reviewed, Postprint (published version)

Deep learning has emerged as a powerful tool for automated object detection, offering unprecedented speed and accuracy in analyzing complex visual data. In the context of marine ecosystem monitoring, convolutional neural networks (CNNs), particularly YOLO-based architectures, have demonstrated remarkable efficiency in detecting and classifying fish species in underwater imagery. Traditional fish identification methods rely on manual annotation, which is both time-consuming and prone to inconsistencies. By implementing a semi-automated labeling approach, where human experts refine AI-generated predictions, the annotation process can be streamlined while ensuring taxonomic precision. A key aspect of this research is the creation of a comprehensive training guide that optimizes the model’s performance by detailing best practices in dataset preparation, annotation techniques, hyperparameter tuning, and augmentation strategies. Using a dataset derived from the OBSEA marine observatory, results indicate that the YOLO extra-large model, trained with a small learning rate and high-resolution images, achieves optimal performance in fish identification. The findings underscore the potential of AI-assisted methodologies in ecological research, offering a scalable and efficient alternative to manual annotation for sustainable marine biodiversity monitoring., This work has been supported by various funding sources and research initiatives. We acknowledge the financial support from grants 2023 INV-2 00044 (position codes 200044TC31 and 200044TC6). Additionally, this research has been funded by the European Commission’s HORIZON-INFRA-2021-SERV-01 program under the iMagine project (grant agreement 101058625). We also recognize the use of the EGI infrastructure with dedicated support from EGI-IFCA-STACK, which contributed to the computational resources required for this study. Furthermore, the researchers wish to acknowledge the support of the Associated Unit Tecnoterra, composed of members from UPC and ICM-CSIC, for their valuable collaboration in this work., Peer Reviewed, Postprint (published version)

This work was developed in the framework of the Research Unit Tecnoterra (ICMCSIC/UPC) and the following project iMagine, funded from the European Union Horizon Europe Programme – Grant Agreement number 101058625. This work benefited from services and resources provided by EGI with dedicated support from IFCA., Peer Reviewed, Postprint (published version)

11th International Workshop on Marine Technology (Martech 2024), 6-7 June 2024, Palma, Balearic Islands, Spain.-- 2 pages, 1 figure, The strategic advancement of monitoring protocols relies on cabled marine observatories for obtaining real-time multiparametric biological and environmental data. The integration of docked mobile platforms, such as underwater crawlers, proves beneficial, extending the surveillance radius of underwater observatories and enhancing their overall performance and functionality. Normally, underwater crawlers are controlled manually, resulting in considerable time and cost expenditures for both crawler operation and monitoring seabed species. To overcome this challenge, the OBSEA Underwater crawler was employed to detect and track a custom object in the laboratory environment. The detection process involved the preparation of a dataset, training a model with YOLO, and finally utilizing an algorithm to track the custom object, This work was developed within the framework of the following project activities: AF was funded by the Pre-doctoral fellowship from AGAUR (ref. BDNS 474817). This work was also partially funded by the Associated Unit Tecnoterra composed by members of Universidad Politécnica de Cataluña (UPC) and the Consejo Superior de Investigaciones Científicas (CSIC), This work received economical support from the European Commission’sHORIZON-INFRA-2021-SERV-01 under the iMagineproject (grant agreement101058625). This work used the EGI infrastructure with the dedicated support of EGI-CESGA-STACK, Peer reviewed

[Data availability]
The in situ data used in this study have been extensively described in Fernández-Mora et al. (2023), which provides detailed information on data accuracy, acquisition time, and the methodology used for data collection and processing. For a comprehensive description of data acquisition and quality, refer to the original reference (Fernández-Mora et al., 2023).
Additionally, the specific in situ data used in this study are available in Fernandez-Mora et al. (2023) and can be accessed and downloaded. This dataset, managed by the Balearic Islands Coastal Observing and Forecasting System (SOCIB), provides valuable records for coastal monitoring. The data are available at the following reference (Fernandez-Mora et al., 2023).
Additionally, all the code used in this study is openly available in the following GitHub repositories:
•
WqSat_Get: A Python package for downloading satellite images (García-Díaz and Aguilar, 2025a).
•
WqSat_Format: A Python package for processing Sentinel-2 and Sentinel-3 images (García-Díaz and Aguilar, 2025b).
•
WqSat_SDB: A Python package for Satellite-Derived Bathymetry (García-Díaz, 2025).
These repositories contain the scripts and methodologies used for data processing, analysis, and visualization, ensuring full transparency and reproducibility of the study., Bathymetry is a fundamental source of information for understanding the marine environment and serves as the primary basis for initiating projects related on marine cartography. Satellite-Derived Bathymetry (SDB) offers a promising alternative to address the lack of current bathymetric data in shallow areas, particularly when utilizing data from satellite platforms such as Sentinel-2, along with advanced Machine Learning and Deep Learning techniques capable of enhancing bathymetric estimation compared to conventional methods. In this study, we evaluated the performance of two different algorithms for estimating SDB in two areas of the Western Mediterranean: a physics-driven model and an Artificial Neural Network (ANN). We assessed the ability of these methods to predict bathymetries over successive years subsequent to algorithm calibration, as well as their capacity to estimate depths of other areas not included in model calibration, thereby evaluating temporal and spatial independence, respectively. In situ depth measurements collected by echo sounders in the study areas were used to train and test the algorithms. Performance metrics, including the coefficient of determination (R2) and Root Mean Square Error (RMSE), consistently yielded results with R2 ≥ 0.8 and RMSE ≤1.5 m across all sections, demonstrating a strong goodness of fit even under scenarios designed to assess temporal and spatial independence. Furthermore, both models showed good performance in estimating both shallow (<10 m) and deeper (>10 m) waters, achieving Median Absolute Error (MedAE) ≤ 0.82 m and Mean Bias Error (MBE) ≤ 0.82 m for shallow waters, and MedAE ≤2.15 m and MBE ≤ 2.33 m even when the depth range extends up to 20 m. The ANN demonstrated a slight improvement in precision, particularly advantageous when a sufficient amount of in situ data is available for algorithm training and validation. Conversely, the physics-driven method, despite being slightly less precise than ANN, offered advantages, especially in study areas with limited data availability., This research was supported by the Grant CNS2023-143630 funded by MICIU/AEI/10.13039/501100011033 and by European Union Next Generation EU/PRTR. This study has been financially supported by the agreement between the Spanish Ministry for Ecological Transition and Demographic Challenge and CSIC, funded by the European Union-Next Generation Program to contribute to the MSFD. The research was also funded by projects from the OAPN (Observatorio TIAMAT, REF: 2715/2021), by the grant FPU20/01294 funded by the Ministry of Science, Innovation and Universities of the Spanish Government, and by the grant PTA2020-018491-I funded by MCIN/AEI/ 10.13039/501100011033 and by “ESF Investing in your future”. This work is co-funded by the iMagine project (Horizon Europe) under Grant number 101058625., Peer reviewed

Images from OBSEA underwater cameras labeled with marine species to train AI-based Object Detection algorithms, European Commission: AI-based image data analysis tools for aquatic research, 101058625, This work contributes to the Institut de Ciències del Mar "Severo Ochoa Centre of Excellence" accreditation CEX2024-001494-S funded by AEI 10.13039/501100011033 of the Spanish Ministry of Science and Innovation, Peer reviewed

The implementations and datasets of use cases are all publicly available at the following addresses:
• The tools available on the iMagine platform as a part of the AI4OS software stack:
https://github.com/ai4os
• The code sources of applications and models implemented through the iMagine use cases:
https://github.com/ai4os-hub
• The ready to use applications of the use cases on the iMagine Marketplace:
https://dashboard.cloud.imagine-ai.eu/marketplace
• The training datasets of the iMagine use cases:
https://zenodo.org/communities/imagine-project/records?q=&f=resource_type%3Adataset, E. Azmi et al., The iMagine project is an EU-funded initiative led by the EGI Foundation. One of the objectives of this project is to provide an AI platform that leverages AI-powered tools to improve the processing and analysis of imaging data from marine and freshwater ecosystems, contributing to the study of the health of oceans, seas, coasts, and inland waters. Connected to the European Open Science Cloud (EOSC), iMagine supports the development, training, and deployment of AI models by collaborating with twelve use cases across diverse aquatic science fields. This collaboration fosters valuable insights and accelerates scientific progress by refining existing solutions in data acquisition, preprocessing, and model deployment. The platform offers trained models as a service, integrating AI tools for image annotation, ensuring the creation of high-quality datasets that comply with FAIR principles. Through these methodologies, iMagine enhances consistency, enabling researchers to efficiently publish and share data in repositories., Beyond its AI tools, iMagine places a strong emphasis on deep learning models, such as convolutional neural networks, for tasks like image classification, object detection, and segmentation, tailored to the unique requirements of aquatic sciences. It also provides robust performance evaluation tools, including experiment tracking, while tackling challenges such as AI model drift and data biases to ensure research reproducibility and transparency. The platform enables users to develop, train, share, and deploy AI models within a flexible environment that integrates with federated cloud and high-performance computing infrastructures, using Docker containers for smooth execution. Additionally, iMagine fosters collaboration with projects like AI4EOSC and Blue-Cloud, and Research Infrastructures such as EMSO and SeaDataNet, expanding its impact on the scientific community., This paper summarizes the key lessons and best practices learned in the iMagine project through the full process of AI-based aquatic image analysis, from data preparation and annotation to model deployment and evaluation. The paper therefore helps aquatic scientists advance their AI-driven image analysis approaches., The project iMagine receives funding from the European Union’s Horizon Europe research and innovation program under grant agreement number 101058625. The AI4OS software stack is developed in the AI4EOSC project, which receives funding from the European Union’s Horizon Europe research and innovation program under grant agreement number 101058593. The use case “Freshwater Diatoms Identification” described in Section 13.10 receives funding from ANR , France (ANR-20-THIA-0010)., Peer reviewed

Images from OBSEA underwater cameras labeled with marine species to train AI-based Object Detection algorithms, The research has been carried out within the framework of the project “Severo Ochoa Center of Excellence” financially supported by ICM-CSIC (CEX2019-000928-S) and the Research Unit Tecnoterra (ICM-CSIC/UPC) of the Spanish Government. This work was supported by the European Union under the Horizon Europe grant 101058625, Peer reviewed

This dataset contains AI-based fish detections at the OBSEA Underwater Observatory (NW Mediterranean sea). The model is a YOLOv8 xlarge trained with labeled underwater pictures from OBSEA, The research has been carried out within the framework of the project “Severo Ochoa Center of Excellence” financially supported by ICM-CSIC (CEX2019-000928-S) and the Research Unit Tecnoterra (ICM-CSIC/UPC) of the Spanish Government. This work was supported by the European Union under the Horizon Europe grant 101058625, The data file in CSV format includes the following columns: - timestamp: Time instance when the picture was taken (in UTC) - SourceImage: URL of the source image - ProcessedImage: URL of the processed image with detections + bounding boxes - SENSOR_ID: Name of the camera that took the picture - Diplodus puntazzo: Number of specimens detected - Oblada melanurus: Number of specimens detected - Parablennius gattorugine: Number of specimens detected - Sarpa salpa: Number of specimens detected - Seriola dumerili: Number of specimens detected - Serranus cabrilla: Number of specimens detected - Sparus aurata: Number of specimens detected - Symphodus mediterraneus: Number of specimens detected - Diplodus sargus: Number of specimens detected - Diplodus vulgaris: Number of specimens detected - Epinephelus costae: Number of specimens detected - Aetomylaeus bovinus: Number of specimens detected - Chromis chromis: Number of specimens detected - Coris julis: Number of specimens detected - Dentex dentex: Number of specimens detected - Diplodus cervinus: Number of specimens detected - Epinephelus marginatus: Number of specimens detected - Mullus surmuletus: Number of specimens detected - Muraena helena: Number of specimens detected - Oblada melanurus: Number of specimens detected - Parablennius gattorugine: Number of specimens detected - Sarpa salpa: Number of specimens detected - Seriola dumerili: Number of specimens detected - Serranus cabrilla: Number of specimens detected - Sparus aurata: Number of specimens detected - Symphodus mediterraneus: Number of specimens detected, Peer reviewed

This dataset contains AI-based fish detections at the OBSEA Underwater Observatory (NW Mediterranean sea). The model is a YOLOv8 xlarge trained with labeled underwater pictures from OBSEA. The model can be found here. The pictures where taken at OBSEA area with cameras pointing to an artifical biotop developed in the framework of the Slagreef project, Ministerio de Ciencia, Innovación y Universidades TED2021-129760B-I00. European Commission Horizon Europe 101058625 iMagine - Imaging data and services for aquatic science, This work contributes to the Institut de Ciències del Mar "Severo Ochoa Centre of Excellence" accreditation CEX2024-001494-S funded by AEI 10.13039/501100011033 of the Spanish Ministry of Science and Innovation, The data file in CSV format includes the following columns: timestamp: Time instance when the picture was taken (in UTC). SourceImage: URL of the source image. ProcessedImage: URL of the processed image with detections + bounding boxes. SENSOR_ID: Name of the camera that took the picture. Diplodus puntazzo: Number of s. Oblada melanurus: Number of specimens detected. Parablennius gattorugine: Number of specimens detected. Sarpa salpa: Number of specimens detected.Seriola dumerili: Number of specimens detected. Serranus cabrilla: Number of specimens detected. Sparus aurata: Number of specimens detected. Symphodus mediterraneus: Number of specimens detected. Diplodus sargus: Number of specimens detected. Diplodus vulgaris: Number of specimens detected. Epinephelus costae: Number of specimens detected. Aetomylaeus bovinus: Number of specimens detected. Chromis chromis: Number of specimens detected. Coris julis: Number of specimens detected. Dentex dentex: Number of specimens detected. Diplodus cervinus: Number of specimens detected. Epinephelus marginatus: Number of specimens detected. Mullus surmuletus: Number of specimens detected. Muraena helena: Number of specimens detected. Oblada melanurus: Number of specimens detected. Parablennius gattorugine: Number of specimens detected. Sarpa salpa: Number of specimens detected. Seriola dumerili: Number of specimens detected. Serranus cabrilla: Number of specimens detected. Sparus aurata: Number of specimens detected. Symphodus mediterraneus: Number of specimens detected, Peer reviewed

Data availability statement: To support reproducibility, transparency, and further research—aligned with the FAIR principles (Wilkinson et al., 2016)—we provide open access to the full image library, the annotated training dataset, and the fully trained classifier model. The FlowCam image collection, containing segmented regions of interest (ROIs) from phytoplankton observations in the Belgian Part of the North Sea (BPNS), is available via the VLIZ Marine Data Archive (Flanders Marine Institute (VLIZ), 2024). The annotated subset used for training has been published on Zenodo LifeWatch observatory data: phytoplankton annotated training set by FlowCam imaging in the Belgian Part of the North Sea (Decrop et al., 2024), together with the final trained classifier (Decrop and Lagaisse, 2025). This accessibility enables other researchers to benchmark and improve upon the classification models, fostering collaboration and accelerating advances in plankton image analysis., Phytoplankton are fundamental components of marine ecosystems and play a critical role in global biogeochemical cycles. Efficient monitoring of marine phytoplankton is crucial for assessing ecosystem health, forecasting harmful algal blooms and sustainable marine management. The integration of high-throughput imaging sensors like FlowCam technology with artificial intelligence (AI) for image recognition has revolutionized phytoplankton monitoring, enabling rapid and accurate class identification. This study introduces an automated image classification workflow designed to improve speed, accuracy and scalability of phytoplankton identification. By leveraging convolutional neural networks (CNNs), the system enhances performance while reducing reliance on traditional, labor-manual identification methods., This work has been funded by the European Union’s Horizon Europe research and innovation programme under grant number 101058625 as part of the iMagine project. It also made use of infrastructure provided by VLIZ, funded by the Research Foundation – Flanders (FWO; Grant I002021N) as part of the Belgian contribution to LifeWatch., Peer reviewed

Deep learning has emerged as a powerful tool for automated object detection, offering unprecedented speed and accuracy in analyzing complex visual data. In the context of marine ecosystem monitoring, convolutional neural networks (CNNs), particularly YOLO-based architectures, have demonstrated remarkable efficiency in detecting and classifying fish species in underwater imagery. Traditional fish identification methods rely on manual annotation, which is both time-consuming and prone to inconsistencies. By implementing a semi-automated labeling approach, where human experts refine AI-generated predictions, the annotation process can be streamlined while ensuring taxonomic precision. A key aspect of this research is the creation of a comprehensive training guide that optimizes the model’s performance by detailing best practices in dataset preparation, annotation techniques, hyperparameter tuning, and augmentation strategies. Using a dataset derived from the OBSEA marine observatory, results indicate that the YOLO extra-large model, trained with a small learning rate and high-resolution images, achieves optimal performance in fish identification. The findings underscore the potential of AI-assisted methodologies in ecological research, offering a scalable and efficient alternative to manual annotation for sustainable marine biodiversity monitoring., This work has been supported by various funding sources and research initiatives. We acknowledge the financial support from grants 2023 INV-2 00044 (position codes 200044TC31 and 200044TC6). Additionally, this research has been funded by the European Commission’s HORIZON-INFRA-2021-SERV-01 program under the iMagine project (grant agreement 101058625). We also recognize the use of the EGI infrastructure with dedicated support from EGI-IFCA-STACK, which contributed to the computational resources required for this study. Furthermore, the researchers wish to acknowledge the support of the Associated Unit Tecnoterra, composed of members from UPC and ICM-CSIC, for their valuable collaboration in this work., Peer Reviewed

This paper addresses the challenges of underwater Visual Odometry (VO) by introducing a novel dataset collected with a monocular camera in a controlled pool environment. The dataset is complemented by preprocessing techniques that mitigate underwater imaging issues such as color distortion and blue-green color cast. A comparative analysis of feature detectors identifies AKAZE as the most effective for underwater
imagery. Additionally, a genetic algorithm (GA) optimizes the RANSAC inlier threshold, enhancing essential matrix estimation. Experimental results highlight a significant reduction in root mean square error (RMSE), with AKAZE achieving a 0.07-meter error compared to traditional methods. This study provides advancements in underwater navigation and mapping, promising improvements for autonomous underwater vehicles (AUVs) and marine research., This work was supported by funding from multiple sources: AF received a
Pre-doctoral fellowship from AGAUR (ref. BDNS 474817) and additional
support from the European Commission’s HORIZON-INFRA-2021-SERV-
01 program through the iMagine project (grant agreement 101058625). The
research also utilized the EGI infrastructure with dedicated support from EGICESGA-
STAC., Peer Reviewed

This work was developed in the framework of the Research Unit Tecnoterra (ICMCSIC/UPC) and the following project iMagine, funded from the European Union Horizon Europe Programme – Grant Agreement number 101058625. This work benefited from services and resources provided by EGI with dedicated support from IFCA., Peer Reviewed

[EN] The iMagine project is an EU-funded initiative led by the EGI Foundation. One of the objectives of this project
is to provide an AI platform that leverages AI-powered tools to improve the processing and analysis of imaging
data from marine and freshwater ecosystems, contributing to the study of the health of oceans, seas, coasts,
and inland waters. Connected to the European Open Science Cloud (EOSC), iMagine supports the development,
training, and deployment of AI models by collaborating with twelve use cases across diverse aquatic science
fields. This collaboration fosters valuable insights and accelerates scientific progress by refining existing
solutions in data acquisition, preprocessing, and model deployment. The platform offers trained models as
a service, integrating AI tools for image annotation, ensuring the creation of high-quality datasets that comply
with FAIR principles. Through these methodologies, iMagine enhances consistency, enabling researchers to
efficiently publish and share data in repositories., The project iMagine receives funding from the European Union s Horizon Europe research and innovation program under grant agreement number 101058625. The AI4OS software stack is developed in the AI4EOSC project, which receives funding from the European Union s Horizon Europe research and innovation program under grant agreement number 101058593. The use case Freshwater Diatoms Identification described in Section 13.10 receives funding from ANR, France (ANR-20-THIA-0010).

[EN] Phytoplankton are fundamental components of marine ecosystems and play a critical role in global biogeochemical cycles. Efficient monitoring of marine phytoplankton is crucial for assessing ecosystem health, forecasting harmful algal blooms and sustainable marine management. The integration of high-throughput imaging sensors like FlowCam technology with artificial intelligence (AI) for image recognition has revolutionized phytoplankton monitoring, enabling rapid and accurate class identification. This study introduces an automated image classification workflow designed to improve speed, accuracy and scalability of phytoplankton identification. By leveraging convolutional neural networks (CNNs), the system enhances performance while reducing reliance on traditional, labor-manual identification methods., The author(s) declare that financial support was received for the
research and/or publication of this article. This work has been
funded by the European Union s Horizon Europe research and
innovation programme under grant number 101058625 as part of
the iMagine project. It also made use of infrastructure provided by
VLIZ, funded by the Research Foundation Flanders (FWO; Grant
I002021N) as part of the Belgian contribution to LifeWatch.

[eng] High-frequency coastal monitoring is essential for understanding shoreline dynamics under climate change and supporting coastal adaptation. Although ground-based remote sensing offers a scalable solution for data acquisition, automated shoreline extraction from beach imaging systems remains challenging due to variable image quality, complex geometries, and environmental conditions. This study presents a robust deep learning framework for instantaneous shoreline extraction from crowd-sourced imagery (i.e., smartphone captures). We benchmarked state-of-the-art segmentation models against a traditional histogram-based deterministic method. Results show that convolutional neural network approaches, particularly DeepLabV3 with 512 ×512 patches, significantly outperform conventional methods, achieving spatial reliability comparable to manual digitisation. Importantly, the model overcomes perspective distortion and interpolation artefacts in rectified imagery across diverse tidal environments, showing strong environmental robustness despite site diversity and data imbalance. The framework has been deployed as an open-source, containerised module within the European Open Science Cloud, enabling transferable and operational image-based coastline monitoring.