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)

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

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