Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
88
datasets available to search
ShareScore release 0.9.0
Dataset results
88 results for “orthomosaic”
UAV-based colour-infrared orthomosaics and digital elevation models of basalts and rock glaciers on Disko Island, West Greenland
<p><span>This data set contains multispectral surveys conducted with an unoccupied aerial vehicle over rock glaciers and steep mafic outcrops (intrusive and flood volcanics) near the coastline of Disko Island.</span></p> <ul> <li><span>Acquisition date: 07.08.2019 – 10.08.2019</span></li> <li><span>Location: Illukunnguaq, Disko Island, Greenland</span></li> <li><span>UAV: SenseFly eBee Plus</span></li> <li><span>Flight altitude above ground level: >100m</span></li> <li><span>Image Overlap forward/side: various</span></li> <li><span>Camera: Parrot Sequoia multispectral</span></li> <li><span>EPSG: 32622</span></li> <li><span>Center coordinates: 69.885277°N, -52.577724°E</span></li> <li><span>Flight mode: automatic flight plan</span></li> </ul> <p><span>Data products: </span></p> <ul> <li><span>Orthomosaic colour-infrared, 10-16 cm pixel resolution</span></li> <li><span>Colour-infrared spectral bands: 790nm, 660nm, 550nm</span></li> <li><span>DEM, 20-30cm pixel resolution</span></li> <li><span>Data coverage: approx. 5500 x 2500 m</span></li> <li><span>Elevation profile: 20-680m </span></li> <li><span>Processing in Agisoft Metashape</span></li> </ul> <p><span>Additional data supplement for article:<br>Barnes, E. (2020). Assessment of Drone-Borne Multispectral Mapping in the Exploration of Magmatic Ni-Cu Sulphides–an Example from Disko Island, West Greenland. <br><em>URN: urn:nbn:se:uu:diva-418858</em></span></p> <p>MULSEDRO field campaign was conducted under scientific survey licence (VU-00158-2019) within mineral exploration licence MEL 2018-16 by Blue Jay Mining PLC. This research has been supported by the project MULSEDRO, funded by HZDR-HIF & EITRawMaterials (project ID 16193) and the European Union.</p>
UAV-based orthomosaic and digital elevation model of a basalt outcrop on Disko Island, West Greenland
<p><span>This data set contains an RGB survey conducted with an unoccupied aerial vehicle (UAV) over a flat basaltic outcrop (intrusive and flood volcanics), surrounded by boreal vegetation (Salix species).</span></p> <ul> <li><span>Acquisition date: 13.08.2019</span></li> <li><span>Location: Qullissat (Qutdlikssat), Disko Island, Greenland</span></li> <li><span>UAV: DJI Mavic 1 Pro</span></li> <li><span>Flight altitude above ground level: 75 m</span></li> <li><span>Image Overlap forward/side: 70 % / 70 %</span></li> <li><span>Camera: RGB</span></li> <li><span>EPSG: 32622</span></li> <li><span>Center coordinates: 70.05330°N, -52.97780°E</span></li> <li><span>Flight mode: manual image acquisition</span></li> </ul> <p><span>Data products: </span></p> <ul> <li><span>Orthomosaic RGB 2.3 cm pixel resolution</span></li> <li><span>DEM 5cm pixel resolution</span></li> <li><span>Processing in Agisoft Metashape</span></li> <li><span>Data coverage: approx. 250 x 360 m</span></li> </ul> <p>Acknowledgements</p> <p><span>MULSEDRO field campaign was conducted under scientific survey licence (VU-00158-2019) within mineral exploration licence MEL 2018-16 by Blue Jay Mining PLC. This research has been supported by the project MULSEDRO, funded by HZDR-HIF & EITRawMaterials (project ID 16193) and the European Union.</span></p>
Topographic data and orthomosaics of the Noordwest Natuurkern project
<p>This data set contains digital elevation models and orthomosaics of the Noordwest Natuurkern region near Bloemendaal aan Zee, the Netherlands. High-frequency wind and rain data are also provided. The data descriptor paper can be found <a href="https://doi.org/10.3390/data9020037" target="_blank" rel="noopener">here</a>, and provides full details of data collection and processing, file format, file naming, and so on. The point clouds from which the elevation models were computed, are not available in this data base but can be obtained from the data creator upon reasonable request.</p> <p>v2.1.0: 41 digital elevation models (20080510 - 20241112; 1x1 m) and 27 orthomosaics (20130501 - 20241112; 1x1 and 0.05x0.05 m) - meta data file - wind and rain data - updated from v2.0.0 on January 8, 2025 (see ChangeLog.txt)</p> <p> </p>
VHR orthomosaic imageries produced by dedicated drone flight campaigns over Lithuania (2021-2022) - Vilnius -AOI 1, Sub-region 1
<p>In the context of the EU-funded project DIONE (No. 870378), targeted drone flight campaigns were conducted in determined Areas of Interest (AOIs) in the Lithuanian territory as a complementary source of information to the spaceborne Earth Observation (EO) open-accessed data. Such an activity attempted to overcome the limited capabilities of EO coarse resolution data to identify small-scale features alongside the agricultural fields, such as the non-productive Ecological Focus Areas (EFAs) and overall to enhance the quality and accuracy (1m or less) of the estimated Land Cover/Land Use elements.</p> <p>The dataset is in fact a collection of drone orthomosaic images delivered in GeoTIFF data format, which were acquired from the drone flight missions over the first subregion of the first AOI located near the capital of Lithuania, (Vilnius). For each area, four flight campaigns were scheduled and conducted on the following dates: </p> <ul> <li><strong>1st flight:</strong> 24-27/5/2021</li> <li><strong>2nd flight:</strong> 26-29/7/2021</li> <li><strong>3rd flight:</strong> 13-16/9/2021</li> <li><strong>4th flight:</strong> 25-28/8/2022</li> </ul>
VHR orthomosaic imageries produced by dedicated drone flight campaigns over Lithuania (2021-2022) - Vilnius -AOI 2
<p>In the context of the EU-funded project DIONE (No. 870378), targeted drone flight campaigns were conducted in determined Areas of Interest (AOIs) in the Lithuanian territory as a complementary source of information to the spaceborne Earth Observation (EO) open-accessed data. Such an activity attempted to overcome the limited capabilities of EO coarse resolution data to identify small-scale features alongside the agricultural fields, such as the non-productive Ecological Focus Areas (EFAs) and overall to enhance the quality and accuracy (1m or less) of the estimated Land Cover/Land Use elements.</p> <p>The dataset is in fact a collection of drone orthomosaic images delivered in GeoTIFF data format, which were acquired from the drone flight missions over the second AOI located near the capital of Lithuania, (Vilnius). For each area, four flight campaigns were scheduled and conducted on the following dates: </p> <ul> <li><strong>1st flight:</strong> 24-27/5/2021</li> <li><strong>2nd flight:</strong> 26-29/7/2021</li> <li><strong>3rd flight:</strong> 13-16/9/2021</li> <li><strong>4th flight:</strong> 25-28/8/2022</li> </ul>
VHR orthomosaic imageries produced by dedicated drone flight campaigns over Cyprus (2021-2022) - Choirokoitia region
<p>In the context of the EU-funded project DIONE (No. 870378), targeted drone flight campaigns were conducted in determined Areas of Interest (AOIs) in the Cypriotic territory as a complementary source of information to the spaceborne Earth Observation (EO) open-accessed data. Such an activity attempted to overcome the limited capabilities of EO coarse resolution data to identify small-scale features alongside the agricultural fields, such as the non-productive Ecological Focus Areas (EFAs) and overall to enhance the quality and accuracy (1m or less) of the estimated Land Cover/Land Use elements.</p> <p>The dataset is in fact a collection of drone orthomosaic images delivered in GeoTIFF data format, which were acquired from the drone flight missions over one of the two determined locations in Cyprus (e.g. Choirokoitia). For each area, four flight campaigns were scheduled and conducted on the following dates: </p> <ul> <li><strong>1st flight:</strong> 29/3/2021- 1/4/2021</li> <li><strong>2nd flight:</strong> 30/8/2021-2/9/2021</li> <li><strong>3rd flight:</strong> 29/11/2021-03/12/2021</li> <li><strong>4th flight:</strong> 17-20/5/2022</li> </ul>
VHR orthomosaic imageries produced by dedicated drone flight campaigns over Cyprus (2021-2022) - Akaki region
<p>In the context of the EU-funded project DIONE (No. 870378), targeted drone flight campaigns were conducted in determined Areas of Interest (AOIs) in the Cypriotic territory as a complementary source of information to the spaceborne Earth Observation (EO) open-accessed data. Such an activity attempted to overcome the limited capabilities of EO coarse resolution data to identify small-scale features alongside the agricultural fields, such as the non-productive Ecological Focus Areas (EFAs) and overall to enhance the quality and accuracy (1m or less) of the estimated Land Cover/Land Use elements.</p> <p>The dataset is in fact a collection of drone orthomosaic images delivered in GeoTIFF data format, which were acquired from the drone flight missions over one of the two determined locations in Cyprus (e.g. Akaki). For each area, four flight campaigns were scheduled and conducted on the following dates: </p> <ul> <li><strong>1st flight:</strong> 29/3/2021- 1/4/2021</li> <li><strong>2nd flight:</strong> 30/8/2021-2/9/2021</li> <li><strong>3rd flight:</strong> 29/11/2021-03/12/2021</li> <li><strong>4th flight:</strong> 17-20/5/2022</li> </ul>
VHR orthomosaic imageries produced by dedicated drone flight campaigns over Lithuania (2021-2022) - Vilnius -AOI 1, Sub-region 2
<p>In the context of the EU-funded project DIONE (No. 870378), targeted drone flight campaigns were conducted in determined Areas of Interest (AOIs) in the Lithuanian territory as a complementary source of information to the spaceborne Earth Observation (EO) open-accessed data. Such an activity attempted to overcome the limited capabilities of EO coarse resolution data to identify small-scale features alongside the agricultural fields, such as the non-productive Ecological Focus Areas (EFAs) and overall to enhance the quality and accuracy (1m or less) of the estimated Land Cover/Land Use elements.</p> <p>The dataset is in fact a collection of drone orthomosaic images delivered in GeoTIFF data format, which were acquired from the drone flight missions over the second subregion of the first AOI located near the capital of Lithuania, (Vilnius). For each area, four flight campaigns were scheduled and conducted on the following dates: </p> <ul> <li><strong>1st flight:</strong> 24-27/5/2021</li> <li><strong>2nd flight:</strong> 26-29/7/2021</li> <li><strong>3rd flight:</strong> 13-16/9/2021</li> <li><strong>4th flight:</strong> 25-28/8/2022</li> </ul>
Kivu Rift 1957-58-59 orthomosaics, version 2020
<p>The orthomosaics provided in this repository are based on historical aerial photographs acquired in 1957, 1958 and 1959 over the Kivu Rift region, in Eastern D.R.Congo, Rwanda and Burundi. This is the 1st version of these orthomosaics, used in the frame of the work of Depicker et al. (2021), on the historical dynamics of landslide risk in the Kivu rift.</p> <p><strong>How to cite the data</strong></p> <p>If you use the orthomosaics, please cite the following to references:</p> <ul> <li>Smets, B., Depicker, A., 2023. Kivu Rift 1957-58-59 orthomosaics, version 2020. <a href="https://doi.org/10.5281/zenodo.7802109">https://doi.org/10.5281/zenodo.7802109</a>.</li> <li>Depicker, A., Jacobs, L., Mboga, N., Smets, B., Van Rompaey, A., Lennert, M., Wolff, E., Kervyn, F., Michellier, C., Dewitte, O., Govers, G., 2021. Historical dynamics of landslide risk from population and forest-cover changes in the Kivu Rift. Nature Sustainability 4, 965-974. <a href="https://doi.org/10.1038/s41893-021-00757-9">https://doi.org/10.1038/s41893-021-00757-9</a>.</li> </ul> <p><strong>Main characteristics of the orthomosaics</strong></p> <ul> <li>Format = GeoTIFF</li> <li>Pixel depth = 16-bit unsigned (unint16)</li> <li>Spatial resolution = 1 or 1.1 m</li> <li>Coordinate system: WGS 84 / UTM 35S (EPSG: 32735)</li> </ul> <p>------------------</p> <p>(c) Royal Museum for Central Africa, 2020-2023</p>
Uncalibrated RGB orthomosaic imagery from UAV campaign at Niwot Ridge, 2017.
Uncalibrated RGB data were collected as part of unmanned aerial vehicle (UAV)/drone campaign during Summer 2017. The purpose of the project was to investigate snow depth variability and spatiotemporal variations and controls on vegetation productivity within the Niwot Ridge LTER Saddle Catchment.
Calibrated Red/Near Infrared orthomosaic imagery from UAV campaign at Niwot Ridge, 2017.
Red/Near Infrared data were collected as part of unmanned aerial vehicle (UAV)/drone campaign during Summer 2017. The purpose of the project was to investigate snow depth variability and spatiotemporal variations and controls on vegetation productivity within the Niwot Ridge LTER Saddle Catchment.
Precision viticulture dataset for detailed vineyard mapping composed of geotagged smartphone ground images, phytosanitary status, UAV orthomosaics, 3D point clouds, and RTK GNSS data - Northern Spain, July 2022
<p>This dataset offers a rich multimodal collection of data from vineyards, designed to enhance agricultural research with a focus on vineyard management and disease monitoring. It includes geotagged smartphone ground images in ".7z" format for detailed plant-level analysis, a ".csv" file detailing plants' phytosanitary status for health assessment, UAV-derived 3D Point Clouds and orthomosaics in ".las" and ".tiff" formats for aerial landscape views, and RTK GNSS data in ".shp" format for precise plant geolocations.</p> <p>This dataset can be combined with other datasets to enable a comprehensive view of the vineyards and improve its value:</p> <div> <ul> <li>Ariza-Sentís, Mar, Sergio Vélez, and João Valente. ‘Dataset on UAV RGB Videos Acquired over a Vineyard Including Bunch Labels for Object Detection and Tracking’. <em>Data in Brief</em> 46 (February 2023): 108848. <a href="https://doi.org/10.1016/j.dib.2022.108848">https://doi.org/10.1016/j.dib.2022.108848</a>.</li> <li>Vélez, Sergio, Mar Ariza-Sentís, and João Valente. ‘VineLiDAR: High-Resolution UAV-LiDAR Vineyard Dataset Acquired over Two Years in Northern Spain.’ <em>Data in Brief</em>, October 2023, 109686. <a href="https://doi.org/10.1016/j.dib.2023.109686">https://doi.org/10.1016/j.dib.2023.109686</a>.</li> <li> <div> <div>Vélez, Sergio, Mar Ariza-Sentís, and João Valente. ‘Dataset on Unmanned Aerial Vehicle Multispectral Images Acquired over a Vineyard Affected by Botrytis Cinerea in Northern Spain’. <em>Data in Brief</em> 46 (February 2023): 108876. <a href="https://doi.org/10.1016/j.dib.2022.108876">https://doi.org/10.1016/j.dib.2022.108876</a>.</div> <div> </div> </div> </li> </ul> </div>
Collaborative UAV-based Orthomosaic of Isimila, Tanzania
<p>An UAV-based orthomosaic of the Middle-Pleistocene archaeological site of Isimila, Tanzania. This version is for the expressed purpose of fostering collaboration of research at the site. GPS coordinates of excavation trenches, surface finds, and other points of interest submitted by any researchers working at the site will be plotted on this regularly updated map. <br> <br> Instructions for submission and contact are available here:<br> <br> https://docs.google.com/document/d/12O3TN7-NuqaszEsiHJ7YUBSztQb1hPfOmaFDy8Zw8BU/edit?usp=sharing</p>
Drone orthomosaics for 'Livestock impacts on an iconic Namib Desert plant are mediated by abiotic conditions'
<p>Drone orthomosaics used in the analyses and figures of: Livestock impacts on an iconic Namib Desert plant are mediated by abiotic conditions, accepted in Oecologia.</p> <p>These drone orthomosaics are a data supplement to the code and data repository here: https://github.com/jtkerb/Nara_Paper_Repo</p>
Processed UAV orthomosaics and DEMs of Fjallsjökull, southeast Iceland (2 of 2)
<p>Orthomosaics of the lower ice surface of Fjallsjökull, southeast Iceland, produced from high resolution UAV surveys undertaken in July 2021. This data was produced as part of my PhD research, and therefore, forms a significant component of my final PhD thesis.</p> <p>All files are in UTM Zone 28N. The resolution of the all the uploaded orthomosaics in this dataset is <strong>0.02 m</strong>. </p> <p>Please note this is dataset 2 out of 2 (due to the 50 GB limit of file uploads). The DEMs and orthomosaics from 2019, DEMs from 2021, as well as three of the orthomosaics from 2021, have been uploaded to dataset 1 of 2 (DOI: 10.5281/zenodo.7105133). </p> <p>For reference, this dataset includes the orthomosaics produced from the following days in July 2021 (in separate files): </p> <p><strong>1)</strong> 8th</p> <p><strong>2) </strong>9th</p> <p><strong>3) </strong>10th</p> <p><strong>4) </strong>11th</p> <p><strong>5)</strong> 12th</p> <p><strong>6)</strong> 15th</p>
RCC Orthomosaics, 2017 - 2019
<p>Visible light orthomosaics of the study reaches in Red Canyon Creek were generated from UAV images and Structure from Motion photogrammetry. The orthomosaics show the study areas one year prior to beaver dam analogue (BDA) installation (2017), immediately after BDA installation (2018), and one year after the BDAs had been constructed (2019). These data are used in Davis et al. (2021), Evaluating the geomorphic channels response to beaver dam analog installation using unoccupied aerial vehicles (https://doi.org/10.1002/esp.5180). The rest of the data used in Davis et al. (2021) is available on CUAHSI HydroShare (http://www.hydroshare.org/resource/7c4ae80863174acfb571ccf8bdac8968).</p>
UAV data of post fire dynamics, Quesenbank, Harz, 2022 (orthomosaics, topography, point clouds)
<p>Unoccupied aerial vehicles (UAVs) were used to investigate a burnt forest site situated in the Harz National Park area near the location Schierke, east of the Brocken, called Quesenbank (<a href="https://www.google.com/maps/place/51%C2%B046'07.6%22N+10%C2%B041'30.2%22E/@51.7682386,10.6905312,489m/data=!3m1!1e3!4m5!3m4!1s0x0:0xd8d7f64a3ff9acf1!8m2!3d51.76877!4d10.69173">51.76877 °N, 10.69173 °E</a>) which is a spruce stand stand severely affected by bark-beetle and windfall. Most of the trees are dead such that they provide high fuel loads for potentially occurring wildfires. The small river Wormke crosses the Quesenbank and separates the survey area between a hiking path and the forest stand.</p> <p>During the 12.08.2022, inhabitants reported a <a href="https://www.ndr.de/nachrichten/niedersachsen/braunschweig_harz_goettingen/Waldbrand-im-Harz-Polizei-geht-von-Brandstiftung-aus,waldbrand882.html">fire</a> near Schierke which was quickly contained by the authorities. Two months after the fire, on 13.10.2022, a team of scientists from GAU Göttingen and TU Berlin was accompanied by a National Park representative for investigation. The site was surveyed by using modern UAVs and by sampling soil strata, ash and vegetation. This report will briefly describe UAV-based surveys and the derived data products.</p> <p>For an overview, see the <strong>report </strong>or download the Maps.zip folder.</p> <p> </p> <p><strong>Acknowledgements</strong></p> <p><strong>Projekt ”Postfeuerdynamik auf Brandflächen im Nationalpark Harz”</strong></p> <p>Dr. Simon Drollinger, Georg-August Universität Göttingen</p> <p>Marlene Düngelhoef, Nationalparkverwaltung Harz</p> <p>Thomas Glinka, Nationalparkverwaltung Harz</p>
Doodleverse/Segmentation Zoo/Seg2Map Res-UNet models for CoastTrain water/other segmentation of RGB 768x768 orthomosaic images
<p><em><strong>Doodleverse/Segmentation Zoo/Seg2Map Res-UNet models for CoastTrain water/other segmentation of RGB 768x768 orthomosaic images</strong></em></p> <p>These Residual-UNet model data are based on Coast Train images and associated labels. https://coasttrain.github.io/CoastTrain/docs/Version%201:%20March%202022/data</p> <p>Models have been created using Segmentation Gym* using the following dataset**: https://doi.org/10.1038/s41597-023-01929-2</p> <p>Image size used by model: 768 x 768 x 3 pixels</p> <p><em>classes:</em><br> 1. Water<br> 2. Other</p> <p><em>File descriptions</em></p> <p>For each model, there are 5 files with the same root name:</p> <p>1. '.json' config file: this is the file that was used by Segmentation Gym* to create the weights file. It contains instructions for how to make the model and the data it used, as well as instructions for how to use the model for prediction. It is a handy wee thing and mastering it means mastering the entire Doodleverse.</p> <p>2. '.h5' weights file: this is the file that was created by the Segmentation Gym* function `train_model.py`. It contains the trained model's parameter weights. It can called by the Segmentation Gym* function `seg_images_in_folder.py`. Models may be ensembled.</p> <p>3. '_modelcard.json' model card file: this is a json file containing fields that collectively describe the model origins, training choices, and dataset that the model is based upon. There is some redundancy between this file and the `config` file (described above) that contains the instructions for the model training and implementation. The model card file is not used by the program but is important metadata so it is important to keep with the other files that collectively make the model and is such is considered part of the model</p> <p>4. '_model_history.npz' model training history file: this numpy archive file contains numpy arrays describing the training and validation losses and metrics. It is created by the Segmentation Gym function `train_model.py`</p> <p>5. '.png' model training loss and mean IoU plot: this png file contains plots of training and validation losses and mean IoU scores during model training. A subset of data inside the .npz file. It is created by the Segmentation Gym function `train_model.py`</p> <p>Additionally, BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU</p> <p><em>References</em><br> *Segmentation Gym: Buscombe, D., & Goldstein, E. B. (2022). A reproducible and reusable pipeline for segmentation of geoscientific imagery. Earth and Space Science, 9, e2022EA002332. https://doi.org/10.1029/2022EA002332 See: https://github.com/Doodleverse/segmentation_gym</p> <p>**Buscombe, D., Wernette, P., Fitzpatrick, S. <em>et al.</em> A 1.2 Billion Pixel Human-Labeled Dataset for Data-Driven Classification of Coastal Environments. <em>Sci Data</em> <strong>10</strong>, 46 (2023). https://doi.org/10.1038/s41597-023-01929-2</p>
Doodleverse/Segmentation Gym SegFormer models for 2-class (wood, other) segmentation of RGB aerial orthomosaic imagery
<p><strong>Doodleverse/Segmentation Gym SegFormer models for 2-class (wood, other) segmentation of RGB aerial orthomosaic imagery</strong></p> <p>This model release is part of the Doodleverse: https://github.com/Doodleverse</p> <p>These Residual-UNet model data are based on RGB (red, green, and blue) images of alluvial river corridors and associated labels. Models are designed to identify subaerial accumulations of large woody debris in orthomosaic imagery. Models have been created using Segmentation Gym* using a dataset of images published here:</p> <p><em>Ritchie, A.C., Curran, C.A., Magirl, C.S., Bountry, J.A., Hilldale, R.C., Randle, T.J., and Duda, J.J., 2018, Data in support of 5-year sediment budget and morphodynamic analysis of Elwha River following dam removals: U.S. Geological Survey data release, https://doi.org/10.5066/F7PG1QWC.</em></p> <p>Classes: {0=other, 1=large woody debris / driftwood}. See https://github.com/Doodleverse for more information about how this model was trained, and how to use it for inference</p> <p><strong>File descriptions</strong></p> <p>1. '.json' config file: this is the file that was used by Segmentation Gym* to create the weights file. It contains instructions for how to make the model and the data it used, as well as instructions for how to use the model for prediction. It is a handy wee thing and mastering it means mastering the entire Doodleverse.</p> <p>2. '.h5' weights file: this is the file that was created by the Segmentation Gym* function `train_model.py`. It contains the trained model's parameter weights. It can called by the Segmentation Gym* function `seg_images_in_folder.py`. Models may be ensembled.</p> <p>3. '_modelcard.json' model card file: this is a json file containing fields that collectively describe the model origins, training choices, and dataset that the model is based upon. There is some redundancy between this file and the `config` file (described above) that contains the instructions for the model training and implementation. The model card file is not used by the program but is important metadata so it is important to keep with the other files that collectively make the model and is such is considered part of the model</p> <p>4. '_model_history.npz' model training history file: this numpy archive file contains numpy arrays describing the training and validation losses and metrics. It is created by the Segmentation Gym function `train_model.py`</p> <p>5. '.png' model training loss and mean IoU plot: this png file contains plots of training and validation losses and mean IoU scores during model training. A subset of data inside the .npz file. It is created by the Segmentation Gym function `train_model.py`</p> <p>Additionally, </p> <p>1. BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU<br> 2. sample_images.zip contains a few example input files, for model testing</p> <p>3. example_validation_outputs.zip contain 50 example validation outputs, consisting of images of ground truth (right) and model output (left). This provides a visually interpretable product to assess model accuracy</p> <p> </p> <p>References</p> <p>*Segmentation Gym: Buscombe, D., & Goldstein, E. B. (2022). A reproducible and reusable pipeline for segmentation of geoscientific imagery. Earth and Space Science, 9, e2022EA002332. https://doi.org/10.1029/2022EA002332 See: https://github.com/Doodleverse/segmentation_gym</p> <p>**<em>Ritchie, A.C., Curran, C.A., Magirl, C.S., Bountry, J.A., Hilldale, R.C., Randle, T.J., and Duda, J.J., 2018, Data in support of 5-year sediment budget and morphodynamic analysis of Elwha River following dam removals: U.S. Geological Survey data release, https://doi.org/10.5066/F7PG1QWC.</em></p>
Doodleverse/Segmentation Gym Residual Unet models for 2-class (alluvial sediment, other) segmentation of RGB aerial orthomosaic imagery
<p><strong>Doodleverse/Segmentation Gym Residual Unet models for 2-class (alluvial sediment, other) segmentation of RGB aerial orthomosaic imagery</strong></p> <p>This model release is part of the Doodleverse: https://github.com/Doodleverse</p> <p>These Residual-UNet model data are based on RGB (red, green, and blue) images of alluvial river corridors and associated labels. Models are designed to identify subaerial alluvial sediment (sand, gravel, etc) in orthomosaic imagery. Models have been created using Segmentation Gym* using a dataset of images published here:</p> <p>Ritchie, A.C., Curran, C.A., Magirl, C.S., Bountry, J.A., Hilldale, R.C., Randle, T.J., and Duda, J.J., 2018, Data in support of 5-year sediment budget and morphodynamic analysis of Elwha River following dam removals: U.S. Geological Survey data release, https://doi.org/10.5066/F7PG1QWC.</p> <p>Classes: {0=other, 1=sediment}. See https://github.com/Doodleverse for more information about how this model was trained, and how to use it for inference</p> <p>File descriptions</p> <p>1. '.json' config file: this is the file that was used by Segmentation Gym* to create the weights file. It contains instructions for how to make the model and the data it used, as well as instructions for how to use the model for prediction. It is a handy wee thing and mastering it means mastering the entire Doodleverse.</p> <p>2. '.h5' weights file: this is the file that was created by the Segmentation Gym* function `train_model.py`. It contains the trained model's parameter weights. It can called by the Segmentation Gym* function `seg_images_in_folder.py`. Models may be ensembled.</p> <p>3. '_modelcard.json' model card file: this is a json file containing fields that collectively describe the model origins, training choices, and dataset that the model is based upon. There is some redundancy between this file and the `config` file (described above) that contains the instructions for the model training and implementation. The model card file is not used by the program but is important metadata so it is important to keep with the other files that collectively make the model and is such is considered part of the model</p> <p>4. '_model_history.npz' model training history file: this numpy archive file contains numpy arrays describing the training and validation losses and metrics. It is created by the Segmentation Gym function `train_model.py`</p> <p>5. '.png' model training loss and mean IoU plot: this png file contains plots of training and validation losses and mean IoU scores during model training. A subset of data inside the .npz file. It is created by the Segmentation Gym function `train_model.py`</p> <p>Additionally, </p> <p>1. BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU<br> 2. sample_images.zip contains a few example input files, for model testing</p> <p>References</p> <p>*Segmentation Gym: Buscombe, D., & Goldstein, E. B. (2022). A reproducible and reusable pipeline for segmentation of geoscientific imagery. Earth and Space Science, 9, e2022EA002332. https://doi.org/10.1029/2022EA002332 See: https://github.com/Doodleverse/segmentation_gym</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.