Find research datasets worth reusing

Underwater images collected by an Autonomous Surface Vehicle in Pointe-Des-Aigrettes, Réunion - 2024-07-08

This dataset was collected by an Autonomous Surface Vehicle in Pointe-Des-Aigrettes, Réunion - 2024-07-08. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 91.26 %, Q2: 2.39 %, Q5: 6.35 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 3.0 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.733 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in Hermitage, Réunion - 2023-11-20

This dataset was collected by an Autonomous Surface Vehicle in Hermitage, Réunion - 2023-11-20. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 83.45 %, Q2: 16.28 %, Q5: 0.27 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.524 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

opencc-by-4.0Jul 2019View details →

Modeling Autonomic Pupillary Responses from External Stimuli Using Machine Learning - Dataset

This page contains the data collected for the paper: Modeling Autonomic Pupillary Responses from External Stimuli Using Machine Learning (<a href="https://doi.org/10.26717/BJSTR.2019.20.003446">DOI:10.26717/BJSTR.2019.20.003446</a>). The dataset consists of spectral and pupillometric data collected during three outdoor/indoor walks. The folders “raw”, “merged”, and “cleaned” contain data collected by the Konica Minolta CL-500A Illuminance Spectrophotometer and Tobii Pro Glasses 2 at three different stages in the data preparation process. The “raw” folder contains uncleaned and unsynchronized .csv/.json files. The “merged” folder contains uncleaned, but synchronized light and ocular data in .csv format. The “cleaned” folder contains a single .csv of cleaned and synchronized data with the derived variables: average pupil diameter and pupil diameter difference.  The best choice of data files will depend on desired analysis. More guidance on how to handle this data can be found in the readMe files located in each subsequent folder. More information on the sensing devices used here can be found at the Minolta and Tobii information links below.  Minolta Information: <a href="https://sensing.konicaminolta.us/uploads/cl-500a_instruction217a_eng-250cl60686.pdf">https://sensing.konicaminolta.us/uploads/cl-500a_instruction217a_eng-250cl60686.pdf</a> Tobii Information: <a href="https://www.tobiipro.com/siteassets/tobii-pro/user-manuals/tobii-pro-glasses-2-user-manual.pdf/?v=1.1.3">https://www.tobiipro.com/siteassets/tobii-pro/user-manuals/tobii-pro-glasses-2-user-manual.pdf/?v=1.1.3</a> The codes used to prepare, analyze, and visualize this data is available in the LightOcular GitHub Repository linked below.  LightOcular GitHub Repo: <a href="https://github.com/mi3nts/LightOcular">https://github.com/mi3nts/LightOcular</a>

opencc-by-4.0Aug 2024View details →

Automated Classification of Conversation Valence and Arousal using Autonomic Nervous System Responses

This repository contains the supplementary file for our study "Automated Classification of Conversation Valence and Arousal using Autonomic Nervous System Responses". The MS Excel file contains all physiological features (individual features and synchrony features) for all valid dyads and all intervals together with self-report ratings of the conversation (Self-Assessment Manikin) and personality trait data (CES-D, BFNES, QCAE). Synchrony features were calculated using code from a previous Zenodo submission (https://zenodo.org/record/7140829).

Underwater images collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10

This dataset was collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 24.43 GB of MP4 files, which were trimmed into 8257 frames (at 2997/1000 fps). The frames are georeferenced. 100.0% of these extracted images are useful and 0.0% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 91.58 %, Q2: 7.65 %, Q5: 0.77 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.299 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 50, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10

This dataset was collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 34.89 GB of MP4 files, which were trimmed into 10570 frames (at 2997/1000 fps). The frames are georeferenced. 100.0% of these extracted images are useful and 0.0% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 69.18 %, Q2: 25.61 %, Q5: 5.22 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://ceruleansonar.com/products/sounder-s500" target="_blank">S500</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.195 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 50, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in Boucan, Réunion - 2023-11-09

This dataset was collected by an Autonomous Surface Vehicle in Boucan, Réunion - 2023-11-09. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 14.54 GB of MP4 files, which were trimmed into 4660 frames (at 2997/1000 fps). The frames are georeferenced. 100.0% of these extracted images are useful and 0.0% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 7.63 %, Q2: 91.59 %, Q5: 0.78 % <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 44, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2024-03-13

This dataset was collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2024-03-13. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 21.93 GB of MP4 files, which were trimmed into 6033 frames (at 2997/1000 fps). The frames are georeferenced. 99.9% of these extracted images are useful and 0.1% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 94.64 %, Q2: 5.34 %, Q5: 0.02 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 10.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.111 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 50, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10

This dataset was collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 21.7 GB of MP4 files, which were trimmed into 6405 frames (at 2997/1000 fps). The frames are georeferenced. 100.0% of these extracted images are useful and 0.0% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 93.23 %, Q2: 5.78 %, Q5: 0.98 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.194 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 44, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in Boucan, Réunion - 2023-11-21

This dataset was collected by an Autonomous Surface Vehicle in Boucan, Réunion - 2023-11-21. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 24.03 GB of MP4 files, which were trimmed into 8608 frames (at 2997/1000 fps). The frames are georeferenced. 97.19% of these extracted images are useful and 2.81% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 91.99 %, Q2: 7.07 %, Q5: 0.94 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.434 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 50, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10

This dataset was collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 29.72 GB of MP4 files, which were trimmed into 7720 frames (at 2997/1000 fps). The frames are georeferenced. 99.77% of these extracted images are useful and 0.23% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 80.53 %, Q2: 14.66 %, Q5: 4.81 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.19 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 44, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10

This dataset was collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 27.62 GB of MP4 files, which were trimmed into 8922 frames (at 2997/1000 fps). The frames are georeferenced. 100.0% of these extracted images are useful and 0.0% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 96.0 %, Q2: 3.92 %, Q5: 0.08 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://ceruleansonar.com/products/sounder-s500" target="_blank">S500</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.444 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 50, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10

This dataset was collected by an Autonomous Surface Vehicle in St-Leu, Réunion - 2023-11-10. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 32.67 GB of MP4 files, which were trimmed into 9381 frames (at 2997/1000 fps). The frames are georeferenced. 98.32% of these extracted images are useful and 1.68% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 97.14 %, Q2: 2.84 %, Q5: 0.03 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://ceruleansonar.com/products/sounder-s500" target="_blank">S500</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.27 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 50, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

Underwater images collected by an Autonomous Surface Vehicle in Boucan, Réunion - 2023-11-22

This dataset was collected by an Autonomous Surface Vehicle in Boucan, Réunion - 2023-11-22. Underwater or aerial images collected by scientists or citizens can have a wide variety of use for science, management, or conservation. These images can be annotated and shared to train IA models which can in turn predict the objects on the images. We provide a set of tools (hardware and software) to collect marine data, predict species or habitat, and provide maps. This dataset is part of larger collection referencing numerous underwater and aerial images <a href="https://doi.org/10.5281/zenodo.11125847" target="_blank">Seatizen Altas</a>. Methods, tools and scientific objectives are also described in a dedicated data paper. <h2>Image acquisition</h2> This session has 16.13 GB of MP4 files, which were trimmed into 6058 frames (at 2997/1000 fps). The frames are georeferenced. 99.75% of these extracted images are useful and 0.25% are useless, according to predictions made by <a href="jacques-v0.1.0_model-20240513_v20.0" target="_blank">Jacques model</a>. Multilabel predictions have been made on useful frames using <a href="https://huggingface.co/lombardata/DinoVdeau-large-2024_04_03-with_data_aug_batch-size32_epochs150_freeze" target="_blank">DinoVd'eau</a> model. <h2> GPS information: </h2> The data was processed with a PPK workflow to achieve centimeter-level GPS accuracy. Base : Files coming from rtk a GPS-fixed station or any static positioning instrument which can provide with correction frames. Device GPS : Emlid Reach M2 Quality of our data - Q1: 94.03 %, Q2: 5.19 %, Q5: 0.78 % <h2> Bathymetry </h2> The data are collected using a single-beam echosounder <a href="https://www.echologger.com/products/single-frequency-echosounder-deep" target="_blank">ETC 400</a>. We only keep the values which have a GPS correction in Q1. We keep the points that are the waypoints. We keep the raw data where depth was estimated between 0.2 m and 50.0 m deep. The data are first referenced against the WGS84 ellipsoid. Then we apply the local geoid if available. At the end of processing, the data are projected into a homogeneous grid to create a raster and a shapefiles. The size of the grid cells is 0.493 m. The raster and shapefiles are generated by linear interpolation. The 3D reconstruction algorithm is ballpivot. <h2>Photogrammetry</h2> OpenDroneMap software was used to create an orthophoto from the raw images. Here is the list of parameters different from the default values for the orthophoto generation. For more details, you can read the log.json file or the 000_photogrammatry_report.pdf report. <code> {'auto_boundary': True, 'cog': True, 'fast_orthophoto': True, 'feature_quality': 'ultra', 'gps_accuracy': 0.1, 'max_concurrency': 44, 'optimize_disk_space': True, 'orthophoto_resolution': 0.1, 'rolling_shutter': True, 'skip_3dmodel': True} </code> <h2> Generic folder structure </h2> YYYYMMDD_COUNTRYCODE-optionalplace_device_session-number ├── DCIM : folder to store videos and photos depending on the media collected. ├── GPS : folder to store any positioning related file. If any kind of correction is possible on files (e.g. Post-Processed Kinematic thanks to rinex data) then the distinction between device data and base data is made. If, on the other hand, only device position data are present and the files cannot be corrected by post-processing techniques (e.g. gpx files), then the distinction between base and device is not made and the files are placed directly at the root of the GPS folder. │ ├── BASE : files coming from rtk station or any static positioning instrument. │ └── DEVICE : files coming from the device. ├── METADATA : folder with general information files about the session. ├── PROCESSED_DATA : contain all the folders needed to store the results of the data processing of the current session. │ ├── BATHY : output folder for bathymetry raw data extracted from mission logs. │ ├── FRAMES : output folder for georeferenced frames extracted from DCIM videos. │ ├── IA : destination folder for image recognition predictions. │ └── PHOTOGRAMMETRY : destination folder for reconstructed models in photogrammetry. └── SENSORS : folder to store files coming from other sources (bathymetry data from the echosounder, log file from the autopilot, mission plan etc.). <h2> Software </h2> All the raw data was processed using our <a href="https://github.com/SeatizenDOI/plancha-workflow/releases/tag/v1.0.3" target="_blank">worflow</a>. All predictions were generated by our <a href="https://github.com/SeatizenDOI/plancha-inference/releases/tag/v1.0.0" target="_blank">inference pipeline</a>. You can find all the necessary scripts to download this data in this <a href="https://github.com/SeatizenDOI/zenodo-tools" target="_blank">repository</a>. Enjoy your data with <a href="https://github.com/SeatizenDOI" target="_blank">SeatizenDOI</a>!

opencc-by-4.0Nov 2022View details →

VHR images were produced in the tunnel tubes of EOAE by autonomous robotic systems (2022)

In the context of the EU-funded project PILOTING (No. 871542), several validation scenarios were scheduled in the three pre-determined pilot sites, i.e. refinery, viaduct, and tunnel, in order to evaluate the good operation of 9 different robotic systems and a versatile data platform. Under this frame, the current dataset was generated during the pilot demonstrations in Metsovo tunnels on 30/9/2022-08/10/2022. The aforementioned were collected by two of the robotic systems, e.g. the CART and the TT-DRONE, covering the execution of the two inspections; a) the general inspection, where the robot covers all the tubes, and b) the local inspection, which captured images in dedicated positions.   This particular dataset is comprised of Very High Resolution (VHR) photos, gathered in the tunnel facilities of EGNATIA ODOS AE (EOAE). During the image capturing, no artificial flashlight was used. The total number of photos acquired from the general inspection is 133 and from the local inspection 8. The data format received by the two robotic systems is in JPEG with the CART vehicle producing images with the approximate dimensions of 9504x6336 pixels, and the TT-DRONE with 6000x4000 pixels.

opencc-by-4.0Nov 2022View details →

PAsCAL WP6 Pilot 2 Autonomous Driving Training

This dataset was collected within the context of the PAsCAL research project between January 2022 and February 2022 at the ACI Vallelunga test circuit and premises in Rome, Italy. Subject of the pilot was a driving training for advanced ADAS systems and test driving of a Level-2+ autonomous vehicle on a test track, performing several different manoeuvres to test the capability of the ADAS systems. Some of the participants were subjected to a driving training for autonomous vehicles before they did the test drive, wherein they had to perform several difficult driving manoeuvres (such as on slippery ground or emergency braking). The purpose of this pilot was to observe whether a driving training improves the driver's capability to use ADAS systems and therefore operate the vehicle in a safer way. Depending on the pilot, it is recommended to adapt existing driving training for beginners, professionals and experienced drivers. In order to analyse the answers given to the questions, it is recommended to consult also the "PAsCAL WP6 Pilots Surveys" dataset, which contains all questions and possible answers.

opengpl-2.0Mar 2023View details →

DeepScenario: An Open Driving Scenario Dataset for Autonomous Driving System Testing

With the rapid development of autonomous driving systems (ADSs), testing ADSs under various driving conditions has become a key method to ensure the successful deployment of ADS in the real-world. However, it is impossible to test all the scenarios due to the inherent complexity and uncertainty of ADSs and the driving tasks. Further, testing of ADSs is expensive regarding time and computational resources. Therefore, a large-scale driving scenario dataset consisting of various driving conditions is needed. To this end, we present an open driving scenario dataset DeepScenario, containing over 30K executable driving scenarios, which are collected by 2880 test executions of three driving scenario generation strategies. Each scenario in the dataset is labeled with six attributes characterizing test results. We further show the attribute statistics and distribution of driving scenarios. For example, there are 1050 collision scenarios, in 917 scenarios there were collisions with other vehicles, 105 and 28 with pedestrians and static obstacles, respectively. This dataset contains: <ol> <li><a href="https://github.com/Simula-COMPLEX/DeepScenario/tree/main/deepscenario-dataset">deepscenario-dataset</a> - DeepScenario dataset, which includes driving scenarios generated by executing three scenario generation strategies: Reinforcement Learning (RL)-based Strategy, Random-based Strategy, Greedy-based Strategy;</li> <li><a href="https://github.com/Simula-COMPLEX/DeepScenario/tree/main/deepscenario-toolset">deepscenario-toolset</a> - The toolset for DeepScenario dataset, including ScenarioCollector that can automatically collect driving scenarios, and ScenarioRunner that can support replaying driving scenarios. We also provide <a href="https://github.com/Simula-COMPLEX/DeepScenario/tree/main/deepscenario-toolset/lgsvl/scenariotoolset">source code</a> and <a href="https://github.com/Simula-COMPLEX/DeepScenario/tree/main/deepscenario-toolset#usage">usage examples</a> for the toolset. </li> </ol> More information about DeepScenario dataset is available in our Github repository: <a href="https://github.com/Simula-COMPLEX/DeepScenario">https://github.com/Simula-COMPLEX/DeepScenario</a>.

opencc-by-4.0Mar 2023View details →

Spineless and overlooked: DNA metabarcoding of autonomous reef monitoring structures reveals intra- and interspecific genetic diversity in Mediterranean invertebrates

Sequence data and stepwise pipeline outputs associated with the article "Spineless and overlooked: DNA metabarcoding of autonomous reef monitoring structures reveals intra- and interspecific genetic diversity in Mediterranean invertebrates". Preprint available here: <a href="https://doi.org/10.22541/au.167085544.47638352/v1">10.22541/au.167085544.47638352/v1</a> Sequence data is deposited in fastq-format in folders by region (Palinuro.tar.gz, Livorno.tar.gz, and Rovinj.tar.gz) and a separate folder for controls (Controls.tar.gz). Each fastq-file contains sequences for a single PCR replicate named by sample and replicate number. Sample names are described in spineless_sample_names.csv. Positive control sequences are described in SM1_positive_controls.csv. Stepwise pipeline outputs are available in the folder Pipeline_outputs_stepwise.zip Scripts used to generate pipeline outputs as well as other aspects of the final article are available at <a href="https://github.com/thomasdotter/spineless-haplotypes">https://github.com/thomasdotter/spineless-haplotypes</a>.