Seatizen Atlas

This deposit offers a comprehensive collection of geospatial and metadata files that constitute the Seatizen Atlas dataset, facilitating the management and analysis of spatial information. <br><br>To navigate through the data, you can use an interface available at <a href="http://seatizenmonitoring.ifremer.re" target="_blank">seatizenmonitoring.ifremer.re</a>, which provides a condensed CSV file tailored to your choice of metadata and the selected area.<br>To retrieve the associated images, you will need to use a script that extracts the relevant frames. A brief tutorial is available here: <a href="https://github.com/SeatizenDOI/zenodo-tools/blob/master/Tutorial.md" target="_blank">Tutorial</a>.<br>All the scripts for processing sessions, creating the geopackage, and generating files can be found here: <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI github repository</a>.<br>All our CSV files are also available in Parquet format.<br><br>The repository includes:<br><ul> <li> <strong> seatizen_atlas_db.gpkg: </strong> geopackage file that stores extensive geospatial data, allowing for efficient management and analysis of spatial information. </li><br> <li> <strong> session_doi.csv: </strong> a CSV file listing all sessions published on Zenodo. This file contains the following columns: </li> <br> <ul> <li> session_name: identifies the session. </li> <li> session_doi: indicates the URL of the session. </li> <li> place: indicates the location of the session. </li> <li> date: indicates the date of the session. </li> <li> raw_data: indicates whether the session contains raw data or not. </li> <li> processed_data: indicates whether the session contains processed data. </li> </ul> <br> <li><strong> metadata_images.csv:</strong> a CSV file describing all metadata for each image published in open access. This file contains the following columns: </li> <br> <ul> <li> OriginalFileName: indicates the original name of the photo. </li> <li> FileName: indicates the name of the photo adapted to the naming convention adopted by the Seatizen team (i.e., YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number_originalimagename). </li> <li> relative_file_path: indicates the path of the image in the deposit. </li> <li> frames_doi: indicates the DOI of the version where the image is located. </li> <li> GPSLatitude: indicates the latitude of the image (if available). </li> <li> GPSLongitude: indicates the longitude of the image (if available). </li> <li> GPSAltitude: indicates the depth of the frame (if available). </li> <li> GPSRoll: indicates the roll of the image (if available). </li> <li> GPSPitch: indicates the pitch of the image (if available). </li> <li> GPSTrack: indicates the track of the image (if available). </li> <li> GPSDatetime: indicates when frames was take (if available). </li> <li> GPSFix: indicates GNSS quality levels (if available). </li> </ul> <br> <li> <strong> metadata_multilabel_predictions.csv:</strong> a CSV file describing all predictions from last multilabel model with georeferenced data. </li><br> <ul> <li> FileName: indicates the name of the photo adapted to the naming convention adopted by the Seatizen team (i.e., YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number_originalimagename). </li> <li> frames_doi: indicates the DOI of the version where the image is located. </li> <li> GPSLatitude: indicates the latitude of the image (if available). </li> <li> GPSLongitude: indicates the longitude of the image (if available). </li> <li> GPSAltitude: indicates the depth of the frame (if available). </li> <li> GPSRoll: indicates the roll of the image (if available). </li> <li> GPSPitch: indicates the pitch of the image (if available). </li> <li> GPSTrack: indicates the track of the image (if available). </li> <li> GPSFix: indicates GNSS quality levels (if available). </li> <li> prediction_doi: refers to a specific AI model prediction on the current image (if available). </li> <li> A column for each class predicted by the AI model. </li> </ul> <br> <li> <strong>metadata_multilabel_annotation.csv: </strong> a CSV file listing the subset of all the images that are annotated, along with their annotations. This file contains the following columns: </li><br> <ul> <li> FileName: indicates the name of the photo. </li> <li> frame_doi: indicates the DOI of the version where the image is located. </li> <li> relative_file_path: indicates the path of the image in the deposit. </li> <li> annotation_date: indicates the date when the image was annotated. </li> <li> A column for each class with values: </li> <ul> <li> 1: if the class is present. </li> <li> 0: if the class is absent. </li> <li> -1: if the class was not annotated. </li> </ul> </ul> <br> <li> <strong>darwincore_multilabel_annotations.zip: </strong> a Darwin Core Archive (DwC-A) file listing the subset of all the images that are annotated, along with their annotations. </li><br></ul><h2> Scientific Publication </h2>If you use this dataset in your research, please consider citing the associated paper: <br><br><pre>@article{contini2025seatizen,<br> title={Seatizen Atlas: a collaborative dataset of underwater and aerial marine imagery},<br> author={Contini, Matteo and Illien, Victor and Julien, Mohan and Ravitchandirane, Mervyn and Russias, Victor <br> and Lazennec, Arthur and Chevrier, Thomas and Rintz, Cam Ly and Carpentier, L{\'e}anne and Gogendeau, Pierre and others}, <br> journal={Scientific Data}, <br> volume={12}, <br> number={1}, <br> pages={67}, <br> year={2025}, <br> publisher={Nature Publishing Group UK London} <br>} <br></pre> <br>For detailed information about the dataset and experimental results, please refer to the previous paper.