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.
51
datasets available to search
ShareScore release 0.9.0
Dataset results
51 results for “aerial images”
Aerial Images_Part 1_Integrating Remote Sensing and Machine Learning for Developing Spatio-Temporal Model to Predict Aquatic Larval Habitats of Malaria
<p>Aerial Images_Part 1_Integrating Remote Sensing and Machine Learning for Developing Spatio-Temporal Model to Predict Aquatic Larval Habitats of Malaria</p>
Lublin 1944 aerial images with spatial overlay index
<p><em><strong>1. Lublin 1944 aerial image overlay index</strong></em> [.geojson or .gml file] is a vectorized, digital form of selected overlay indexes for degree square 51N022E (https://catalog.archives.gov/id/44241929) of German Flown Aerial Photographs,1939-1945 (https://catalog.archives.gov/id/306065) archived in National Archives and Records Administration, College Park, MD.</p> <p>2. The vectorized index contains geometries, attributes and other metadata of 104 aerial images of Lublin [Poland] captured by Luftwaffe reconaissance from 10th May 1944 to 6th December 1944.</p> <p>3. The dataset contains the archive of 104 digital copies of aerial images, scaned with A2-3050-Sharp363N. The images are .jpg files with 24-bit colour depth and resolution 600 dpi. File size: from 7,5 MB to 20 MB.</p> <p>4. The mosaic of aerial images is uploaded in Ortofotomapa_1944_modificado_3.tif - 0,8GB file. TFW, AUX and OVR files added for GIS users. TPK file with ESRI tiled package is added. This is also available via spatial data services (TMS and WMTS):</p> <p>- XYZ/TMS layers available at <strong><em>//ortolub.umcs.pl/data/tiles_3857/{z}/{x}/{y}.png </em></strong>or via https://ortolub.umcs.pl/map_en.html</p> <p>- WMTS layer available at <strong><em>//tiles.arcgis.com/tiles/STaxETJ8DGWEoQ8D/arcgis/rest/services/Ortolub_1944/MapServer </em></strong>or via ArcGIS Online https://www.arcgis.com/home/item.html?id=d5dd97b49b014d6ea0e6bc221fc37668</p> <p>5. The mosaic cropped to 1931-1947 city boundaries are added: Lublin_1944_aerial_10k_600dpi_gsc.jpg and Lublin_1944_aerial_adm_10k_600dpi_gsc.jpg with area outside the boundaries masked. This is also available at Wikimedia Commons, https://commons.wikimedia.org/wiki/File:Lublin_1944_aerial_image.jpg</p> <p>6. The project and the platform <em><strong>https://ortolub.umcs.pl</strong></em> was developed under the Polish National Science Centre grant programme - Miniatura 4.0. ref. no. 2020/04/X/HS4/00382. I hereby share my work under Creative Commons license CC BY-SA 4.0 (Attribution - ShareAlike).</p>
Near real-time ultrahigh-resolution imaging from unmanned aerial vehicles for sustainable land use management and biodiversity conservation in semi-arid savanna under regional and global change (SAVMAP)
<p>To prevent aggravation of existing poverty in semi-arid savannas, a comprehensive concept for the sustainable adaptive management and use of these ecosystems under unprecedented conditions is needed. SAVMAP is an innovative, trans-, and inter-disciplinary initiative whose goal is to develop a valuable monitoring tool for both sustainable land-use management and rare species conservation (black rhinoceros) in semi-arid savanna in Namibia. SAVMAP uses near real-time ultrahigh-resolution photographic imaging (NURI) facilitated by unmanned aerial vehicles (UAVs) designed at EPFL.</p>
AIDERv2 (Aerial Image Dataset for Emergency Response Applications)
<div>SUMMARY OF DATASET</div> <div> </div> <div>• This dataset consist of 167723 aerial images divided into 4 classes.</div> <div> </div> <div>• The dataset contains three commonly occurring natural disasters</div> <div>earthquake/collapsed buildings, flood, wildfire/fire, and a normal class; do not reflect any disaster</div> <div> </div> <div>• The images can be loaded as numpy arrays using Python programming language and then used to train a Convolutional Neural Network to detect natural disasters from aerial images.</div> <div> </div> <div>• The images are resized to 224x224x3 (heighty,width,channel number) when loaded as numpy arrays.</div> <div> </div> <div>• The dataset is an extension of the AIDER dataset (Aerial Image Dataset for Emergency Response Applications). </div> <div> </div> <div>• Additional images were collected by open source databases and extracted images as frames of videos downloaded from YouTube. </div> <div> </div> <div> </div> <div>The table below shows the number of images in each set.</div> <div> </div> <div> Train Validation Test Total</div> <div> Earthquakes 1927 239 239 2405</div> <div> Floods 4063 505 502 5070</div> <div> Fire 3509 439 436 4384</div> <div> Normal 3900 487 477 4864</div> <div> Total 13399 1670 1654 16723</div> <div> </div> <div> </div> <div>If you use this dataset please cite the following publications:</div> <div> </div> <div>[1] Shianios, D., Kyrkou, C., Kolios, P.S. (2023). A Benchmark and Investigation of Deep-Learning-Based Techniques for Detecting Natural Disasters in Aerial Images. In: Tsapatsoulis, N., <em>et al.</em> Computer Analysis of Images and Patterns. CAIP 2023. Lecture Notes in Computer Science, vol 14185. Springer, Cham. https://doi.org/10.1007/978-3-031-44240-7_24</div> <div>Link: https://link.springer.com/chapter/10.1007/978-3-031-44240-7_24</div> <div> </div> <div>[2] D. Shianios, P. Kolios, C. Kyrkou, "DiRecNetV2: A Transformer-Enhanced Network for Aerial Disaster Recognition", SN Computer Science, 2024 (Accepted to Appear)</div> <div> </div> <div> </div> <div> </div> <div> </div> <div>DATASET FOLDERS FORMAT</div> <div> </div> <div>└───data</div> <div>│ │</div> <div>│ └───Dataset_Images</div> <div>│ │ └───Train</div> <div>│ │ │ | └───Earthquake</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Flood</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Normal</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Wildfire</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ └───Val</div> <div>│ │ │ | └───Earthquake</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Flood</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Normal</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Wildfire</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ └───Test</div> <div>│ │ │ | └───Earthquake</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Flood</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Normal</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div>│ │ │ | └───Wildfire</div> <div>│ │ │ | img (1).jpg</div> <div>│ │ │ | img (2).jpg</div> <div>│ │ │ | .....</div> <div> </div> <div> </div> <div> </div> <div> </div> <div> </div> <div>DATA SOURCES AND DATA COLLECTION</div> <div> </div> <div>OPEN SOURCE DATABASES</div> <div> </div> <div>└───AIDER </div> <div>https://zenodo.org/record/3888300#.Yuu11nZBxD-</div> <div>Kyrkou, C. and Theocharides, T., 2020. EmergencyNet: Efficient aerial image classification for drone-based emergency monitoring using atrous convolutional feature fusion. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 13, pp.1687-1699.</div> <div> </div> <div>└───ERA </div> <div>https://lcmou.github.io/ERA_Dataset/</div> <div>Mou, L., Hua, Y., Jin, P. and Zhu, X.X., 2020. Era: A data set and deep learning benchmark for event recognition in aerial videos [software and data sets]. IEEE Geoscience and Remote Sensing Magazine, 8(4), pp.125-133.</div> <div>@article{eradataset,</div> <div> title = {{ERA: A dataset and deep learning benchmark for event recognition in aerial videos}},</div> <div> author = {Mou, L. and Hua, Y. and Jin, P. and Zhu, X. X.},</div> <div> journal = {IEEE Geoscience and Remote Sensing Magazine},</div> <div> year = {in press}</div> <div>}</div> <div> </div> <div> </div> <div> </div> <div>└───ISBDA</div> <div>https://drive.google.com/file/d/1kEKJ8kr1aScXz_1El7Mn-Yi0ANducQIW/view</div> <div>Zhu, X., Liang, J. and Hauptmann, A., 2021. Msnet: A multilevel instance segmentation network for natural disaster damage assessment in aerial videos. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (pp. 2023-2032).</div> <div>@misc{zhu2020msnet,</div> <div> title={MSNet: A Multilevel Instance Segmentation Network for Natural Disaster Damage Assessment in Aerial Videos},</div> <div> author={Xiaoyu Zhu and Junwei Liang and Alexander Hauptmann},</div> <div> year={2020},</div> <div> eprint={2006.16479},</div> <div> archivePrefix={arXiv},</div> <div> primaryClass={cs.CV}</div> <div>}</div> <div> </div> <div> </div> <div>└───Floods 2013</div> <div>https://github.com/cvjena/eu-flood-dataset</div> <div>Barz, B., Schröter, K., Münch, M., Yang, B., Unger, A., Dransch, D. and Denzler, J., 2019. Enhancing flood impact analysis using interactive retrieval of social media images. arXiv preprint arXiv:1908.03361.</div> <div>@article{barz2019enhancing,</div> <div> title={Enhancing flood impact analysis using interactive retrieval of social media images},</div> <div> author={Barz, Bj{\"o}rn and Schr{\"o}ter, Kai and M{\"u}nch, Moritz and Yang, Bin and Unger, Andrea and Dransch, Doris and Denzler, Joachim},</div> <div> journal={arXiv preprint arXiv:1908.03361},</div> <div> year={2019}</div> <div>}</div> <div> </div> <div>└───Wildfire Research</div> <div>http://wildfire.fesb.hr/index.php?option=com_content&view=article&id=58&Itemid=54</div> <div> </div> <div> </div> <div>└───PyImages</div> <div>https://drive.google.com/file/d/1NvTyhUsrFbL91E10EPm38IjoCg6E2c6q/view</div> <div>The dataset was curated by PyImageSearch reader, Gautam Kumar.</div> <div> </div> <div> </div> <div> </div> <div> </div> <div>YOUTUBE VIDEOS</div> <div> </div> <div>└───Collapsed Buildings/Earthquakes</div> <div>• https://www.youtube.com/watch?v=TMow3WPcZrQ&t=133s&ab_channel=GORKHALYFOUNDATION</div> <div>• https://www.youtube.com/watch?v=_HT0tYKKjBI&t=47s&ab_channel=Effect.org</div> <div>• https://www.youtube.com/watch?v=rkb3y6K3waU</div> <div>• https://www.youtube.com/watch?v=yir6ArRZY4o&t=109s&ab_channel=UnicefUK</div> <div>• https://www.youtube.com/watch?v=CM9APmIR9Fk&ab_channel=ToonsZilla</div> <div>• https://www.youtube.com/watch?v=tmx2w6drAeU&ab_channel=AssociatedPress</div> <div>• https://www.youtube.com/watch?v=kuSEe8Emwrk&ab_channel=BloombergQuicktake%3ANow</div> <div>• https://www.youtube.com/watch?v=qoFHA3-m5ag&ab_channel=NBCNews</div> <div>• https://www.youtube.com/watch?v=MM3PToqEPhQ&ab_channel=GuardianNews</div> <div>• https://www.youtube.com/watch?v=zB_-TRnGuZE&ab_channel=DISASTERNEWS</div> <div>• https://www.youtube.com/watch?v=TqAMQQOEsBs&ab_channel=WHAS11</div> <div>• https://www.youtube.com/watch?v=0ixjTt-jmok&ab_channel=EveningStandard</div> <div>• https://www.youtube.com/watch?v=bNGA8Ms3d70&ab_channel=CatersClips</div> <div>• https://www.youtube.com/watch?v=wJ-2d5t23Lg&ab_channel=DailyDose</div> <div>• https://www.youtube.com/watch?v=ewUcI7I6Gf4&ab_channel=NBCNews</div> <div>• https://www.youtube.com/watch?v=Wx1cjOdlMZ4&ab_channel=ABCNews%28Australia%29</div> <div>• https://www.youtube.com/watch?v=jiMK_sVmbXk&t=12s&ab_channel=NewChinaTV </div> <div>• https://www.youtube.com/watch?v=M9au_9A2YRo&ab_channel=GuardianNews</div> <div>• https://www.youtube.com/watch?v=i6Lh8IXPjso&ab_channel=TheSun</div> <div>• https://www.youtube.com/watch?v=CKwxEr3I4Y8&ab_channel=GuardianNews</div> <div>• https://www.youtube.com/watch?v=hxqzcajBCNg&list=RDCMUCD3KREyo3IqCLBC-4khGgIw&index=3&ab_channel=WXChasing</div> <div>• https://www.youtube.com/watch?v=2GEeTDuf9mI&list=RDCMUCD3KREyo3IqCLBC-4khGgIw&index=6&ab_channel=WXChasing</div> <div>• https://www.youtube.com/watch?v=bDOuZWxIyNQ&list=RDCMUCD3KREyo3IqCLBC-4khGgIw&index=9&ab_channel=WXChasing</div> <div>• https://www.youtube.com/watch?v=vzoSADijLCQ&list=RDCMUCD3KREyo3IqCLBC-4khGgIw&index=15&ab_channel=WXChasing</div> <div>• https://www.youtube.com/watch?v=ZaL1fldTEAk&list=RDCMUCD3KREyo3IqCLBC-4khGgIw&index=17&ab_channel=WXChasing</div> <div>• https://www.youtube.com/watch?v=QSV81FdilZE&ab_channel=GlobalNews</div> <div>• https://www.youtube.com/watch?v=KgOk9otW1Bg&ab_channel=EricFeijten</div> <div> </div> <div> </div> <div> </div> <div>└───Floods</div> <div>• https://www.youtube.com/watch?v=DJqgv8Sa5bA&t=317s&ab_channel=7NEWSAustralia</div> <div>• https://www.youtube.com/watch?v=w5FintiCLJU&t=9s&ab_channel=GuardianNews</div> <div>• https://www.youtube.com/watch?v=HjMymNN6Ajc&t=143s&ab_channel=BioLogicTreeServices</div> <div>• https://www.youtube.com/watch?v=Tmba18C94C8&ab_channel=AL.com</div> <div>• https://www.youtube.com/watch?v=Dqvpv4Vg4lk&t=63s&ab_channel=ElevenEleven</div> <div>• https://www.youtube.com/watch?v=N7QGicNtN2A&ab_channel=PKSVideoProductions</div> <div>• https://www.youtube.com/watch?v=8CHagyQG16Q&ab_channel=Stolly-Sven</div> <div>• https://www.youtube.com/watch?v=vjH3zFqdzcE&ab_channel=BenChilders</div> <div>• https://www.youtube.com/watch?v=GFw89UB4fE8&ab_channel=BenChilders</div> <div>• https://www.youtube.com/watch?v=heP3LEJ_NkE&ab_channel=7NEWSAustralia</div> <div> </div> <div> </div> <div> </div> <div>└───Fires</div> <div>• https://www.youtube.com/watch?v=gbM_NPx2GPc&t=201s&ab_channel=WXChasing</div> <div>• https://www.youtube.com/watch?v=M97sJdyeEM4&t=72s&ab_channel=Sanuck176</div> <div>• https://www.youtube.com/watch?v=1Z2K6lDt76M&t=557s&ab_channel=TheRelaxationChannel</div> <div> </div> <div>└───Normal</div> <div>• https://www.youtube.com/watch?v=SyxjsuNHWhM&t=328s&ab_channel=OneManWolfPack</div> <div>• https://www.youtube.com/watch?v=f1PTWsBtrtc&ab_channel=ChernobylPug</div> <div> </div> <div> </div> <div> </div> <div> </div> <div> </div>
Table 4 in A comparison of image and observer based aerial surveys of narwhal
<p><i>Table 4.</i> Summary statistics of survey results and narwhal abundance estimates from surveys conducted in Melville Bay, West Greenland from 25 to 30 August 2014. Off-effort sightings by the aerial observers were only included in modeling of the detection function. Strata without narwhal sightings were excluded from this table (<i>i.e.</i>, south and northwest). Abundance estimates are corrected for availability bias with <i>â</i> (0) being 0.257 for the aerial observer sightings and 0.22 for the image sightings. Aerial observer sightings were truncated at 1,300 m. CV values are given in parentheses. Ind. = individuals.</p><table><tbody><tr><th></th><th></th><th></th><th></th><th></th><th>Uncorrected</th><th></th><th></th><th>Abundance</th></tr></tbody><tbody><tr><th></th><td></td><td></td><td>Encounter</td><td>Mean</td><td>density</td><td>Uncorrected</td><td>Uncorrected</td><td>of individuals</td></tr><tr><th>Platform</th><td>Stratum</td><td>Encounter rate (groups/km)</td><td>rate (ind./km)</td><td>group size</td><td>of groups (groups/km2)</td><td>density of ind. (ind./km2)</td><td>abundance of individuals</td><td>corrected for availability bias</td></tr><tr><th>Aerial</th><td>Northeast</td><td>0.032 (0.86)</td><td>0.084 (0.78)</td><td>2.6 (0.20)</td><td>0.046 (0.93)</td><td>0.115 (0.90)</td><td>300.7 (0.89)</td><td>1,171.0 (0.90)</td></tr><tr><th>observers</th><td>Central</td><td>0.068 (0.50)</td><td>0.261(0.53)</td><td>3.8 (0.11)</td><td>0.046 (0.53)</td><td>0.177 (0.52)</td><td>366.5 (0.52)</td><td>1,426.1 (0.53)</td></tr><tr><th></th><td>All strata</td><td>0.053 (0.41)</td><td>0.187(0.44)</td><td>3.5 (0.10)</td><td>0.046 (0.58)</td><td>0.142 (0.50)</td><td>667.2 (0.50)</td><td>2,596.1 (0.51)</td></tr><tr><th>Images</th><td>Northeast</td><td>0.049 (0.94)</td><td>0.105(0.92)</td><td>2.1 (0.91)</td><td>0.049 (0.94)</td><td>0.105 (0.92)</td><td>217.8 (0.92)</td><td>990 (0.93)</td></tr><tr><th></th><td>Central</td><td>0.059 (0.70)</td><td>0.130(0.61)</td><td>2.2 (0.71)</td><td>0.059 (0.70)</td><td>0.130 (0.61)</td><td>340.0 (0.61)</td><td>1,545.5 (0.61)</td></tr><tr><th></th><td>All strata</td><td>0.055 (0.55)</td><td>0.119(0.50)</td><td>2.2 (0.78)</td><td>0.055 (0.55)</td><td>0.119 (0.50)</td><td>557.8 (0.50)</td><td>2,535.5 (0.51)</td></tr></tbody></table>
Table 2 in A comparison of image and observer based aerial surveys of narwhal
<p><i>Table 2.</i> Summary of sightings by image analysts and/or aerial observers, with and without truncation at 500 m. Where both aerial observer pairs had the same sighting with different group size or perpendicular distance estimates, an averaged value was used.</p><table><tbody><tr><th></th><th></th><th></th><th>Only</th><th></th><th>Sighted by</th><th>Sighted by</th></tr></tbody><tbody><tr><th></th><td>Sighted by</td><td></td><td>sight by</td><td>Only</td><td>image</td><td>image analyst</td></tr><tr><th></th><td>image</td><td>Sighted by</td><td>image</td><td>sighted by</td><td>analyst</td><td>and/or</td></tr><tr><th></th><td>analyst</td><td>observer</td><td>analyst</td><td>observer</td><td>and observer</td><td>observer</td></tr><tr><th>No truncation</th></tr><tr><th>No. of groups</th><td>62</td><td>63</td><td>34</td><td>35</td><td>28</td><td>97</td></tr><tr><th>No. of animals</th><td>135</td><td>227</td><td>62</td><td>124</td><td>88</td><td>274</td></tr><tr><th>Average group size</th><td>2.2</td><td>3.6</td><td>1.8</td><td>3.5</td><td>3.1</td><td>2.8</td></tr><tr><th>Average distance</th><td>227</td><td>580</td><td>237</td><td>839</td><td>329</td><td>454</td></tr><tr><th>Truncation = 500 m</th><td></td><td></td><td></td><td></td><td></td><td></td></tr><tr><th>No. of groups</th><td>62</td><td>36</td><td>34</td><td>8</td><td>28</td><td>70</td></tr><tr><th>No. of animals</th><td>135</td><td>126</td><td>62</td><td>24</td><td>88</td><td>173</td></tr><tr><th>Average group size</th><td>2.2</td><td>3.5</td><td>1.8</td><td>3.0</td><td>3.1</td><td>2.5</td></tr><tr><th>Average distance</th><td>227</td><td>269</td><td>237</td><td>329</td><td>234</td><td>247</td></tr></tbody></table>
Table 3 in A comparison of image and observer based aerial surveys of narwhal
<p><i>Table 3.</i> AIC values after fitting explanatory variables to the DS and MR models. The final model chosen, Model 1, is given in bold and ΔAIC indicates the difference between the chosen model and the specified model. HN indicates a half-normal form and HR indicates a hazard-rate form for the DS model. The explanatory variables are perpendicular distance (D), group size (S), group size as a factor with three classes (1, 2–5, and <i>≥</i> 6 narwhals) (S3), Beaufort (BF), side of airplane (SP), observer pair (O), and time to next observation <i>≤</i> 10 s (T). Appendix S3 provides an overview of all the models that were developed.</p><table><tbody><tr><th>Model</th><th>DS model</th><th>MR model</th><th>No. of parameters</th><th>AIC</th><th>ΔAIC</th></tr></tbody><tbody><tr><th><b>1</b></th><td><b>HN: D + BF</b></td><td><b>D + O + S</b> <b>3</b></td><td><b>5</b></td><td><b>1,330.91</b></td><td><b>0</b></td></tr><tr><th>2</th><td>HN: D + BF</td><td>D + O + S</td><td>5</td><td>1,332.81</td><td>1.90</td></tr><tr><th>3</th><td>HN: D + BF + S3</td><td>D + O + T</td><td>6</td><td>1,333.10</td><td>2.19</td></tr><tr><th>4</th><td>HN: D + BF</td><td>D + O</td><td>4</td><td>1,333.50</td><td>2.59</td></tr><tr><th>5</th><td>HN: D + BF + S3</td><td>D + O</td><td>5</td><td>1,335.08</td><td>4.17</td></tr><tr><th>6</th><td>HR: D + BF</td><td>D + O + S3</td><td>5</td><td>1,336.97</td><td>6.06</td></tr></tbody></table>
Table 1 in A comparison of image and observer based aerial surveys of narwhal
<p><i>Table 1.</i> Summary of survey effort in the four strata during aerial surveys conducted in Melville Bay from 25 to 30 August 2014.</p><table><tbody><tr><th></th><th></th><th></th><th></th><th></th><th></th><th></th><th></th><th></th><th></th><th></th><th>Aerial observers</th></tr></tbody><tbody><tr><th></th><td></td><td></td><td></td><td></td><td></td><td>Image</td><td></td><td>Aerial observers</td><td>(truncated 500 m)</td></tr><tr><th></th><td></td><td>Planned/</td><td>Length of</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr><tr><th></th><td>Area</td><td>completed</td><td>surveyed</td><td>Analyzed</td><td>No. of</td><td>No.</td><td>Ave. group</td><td>No. of</td><td>No.</td><td>Ave. group</td><td>No. of</td><td>No.</td><td>Ave. group</td></tr><tr><th>Stratum</th><td>(km2)</td><td>transects (no.)</td><td>transects (km)</td><td>images a</td><td>sightings</td><td>of ind.b</td><td>size <i>n</i> (CV)</td><td>sightings</td><td>of ind.</td><td>size <i>n</i> (CV)</td><td>sightings</td><td>of ind.</td><td>size <i>n</i> (CV)</td></tr><tr><th>NW</th><td>6,376</td><td>5 / 4</td><td>226</td><td>2,066</td><td>0</td><td>0</td><td>0 (0)</td><td>0</td><td>0</td><td>0 (0)</td><td>0</td><td>0</td><td>0 (0)</td></tr><tr><th>NE</th><td>2,076</td><td>11 / 12</td><td>468</td><td>6,572</td><td>23</td><td>49</td><td>2.1 (0.95)</td><td>15</td><td>39</td><td>2.6 (0.77)</td><td>13</td><td>32.5</td><td>2.5 (0.76)</td></tr><tr><th>C</th><td>2,621</td><td>9 / 18</td><td>663</td><td>8,193</td><td>39</td><td>86</td><td>2.2 (0.73)</td><td>48</td><td>188</td><td>3.9 (0.69)</td><td>23</td><td>93.5</td><td>4.1 (0.61)</td></tr><tr><th>S</th><td>3,748</td><td>13 / 10</td><td>575</td><td>4,105</td><td>0</td><td>0</td><td>0 (0)</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr><tr><th>All strata</th><td>14,821</td><td>38 / 44</td><td>1,932</td><td>20,936</td><td>62</td><td>135</td><td>2.2 (0.77)</td><td>63</td><td>227</td><td>3.6 (0.72)</td><td>36</td><td>126</td><td>3.5 (0.69)</td></tr></tbody></table><p><sup>a</sup> Including 180 off-effort images that were not used in the analysis. <sup>b</sup> Ind. = individuals.</p>
Aerial images of sea ice during the Oden_AO2018 campaign (Arctic Ocean 2018: MOCCHA - ACAS - ICE) Central Arctic / North Pole drift station in 2018 on IB ODEN
<p>35 oblique aerial images of sea ice (.jpg) were obtained during helicopter and drone flights during the Oden_AO2018 campaign (Arctic Ocean 2018: MOCCHA - ACAS - ICE) Central Arctic / North Pole drift station in 2018 on IB ODEN. The images display the drifting, melt-pond covered multi-year ice (MYI) floe close to the geographic North Pole in autumn 2018 on which the IB ODEN was anchored to. The images were taken on 14, 16, 28 August and 13 September. The images document the evolution of melt ponds between their fully developed stage in late summer towards their autumn characteristics with a refrozen surface and finally a snow cover.</p> <p>For details, please see the respective publication Anhaus, P., Katlein, C., Nicolaus, M., Hoppmann, M., and Haas, C.: From Bright Windows to Dark Spots: The Evolution of Melt Pond Optical Properties during Refreezing. Currently under review in Geophysical Research Letters. Preprint available <a href="https://doi.org/10.1002/essoar.10507628.1">doi:10.1002/essoar.10507628.1</a></p> <p>For the cruise report, please see Leck, C., Matrai, P. A., Perttu, A.-M., and Gårdfeldt, K. (2019): Expedition report: SWEDARTIC Arctic Ocean 2018. Luleå: Swedish Polar Research Secretariat. <a href="https://nbn-resolving.org/urn:nbn:se:polar:diva-8405">urn:nbn:se:polar:diva-8405</a></p>
Reconstructed Aneto glacier surfaces from historic aerial image photogrammetry (1981) and remote sensing techniques (2020, 2021, 2022)
<p>The Aneto Glacier, is the largest glacier in the Pyrenees. Its shrinkage and wastage have been continuous in recent decades, and there are signs of accelerated melting in recent years. In this study, changes in the surface and ice thickness of the Aneto Glacier from 1981 to 2022 are investigated using historical aerial imagery, airborne LiDAR point clouds, and UAV imagery. A GPR survey conducted in 2020, combined with data from photogrammetric analyses, allowed us to reconstruct the current ice thickness and also the existing ice distribution in 1981 and 2011. Over the last 41 years, the total glaciated area has shrunk by 64.7% and the ice thickness has decreased, on average, by 30.5 m. The mean remaining ice thickness in autumn 2022 was 11.9 m, as against the mean thicknesses of 32.9 m, 19.2 m reconstructed for 1981 and 2011 and 15.0 m observed in 2020 respectively. The results demonstrate the critical situation of the glacier, with an imminent segmentation into two smaller ice bodies and no evidence of an accumulation zone. We also found that the occurrence of an extremely hot and dry year, as observed in the 2021–2022 season, leads to a drastic degradation of the glacier, posing a high risk to the persistence of the Aneto Glacier, a situation that could extend to the rest of the Pyrenean glaciers in a relatively short time. </p>
Fusion of Single and Integral Multispectral Aerial Images
<p>Abstract: <span>An adequate fusion of the most significant salient information from multiple input channels is essential for many aerial imaging tasks. While multispectral recordings reveal features in various spectral ranges, synthetic aperture sensing makes occluded features visible. We present a first and hybrid (model- and learning-based) architecture for fusing the most significant features from conventional aerial images with the ones from integral aerial images that are the result of synthetic aperture sensing for removing occlusion. It combines the environment’s spatial references with features of unoccluded targets that would normally be hidden by dense vegetation. Our method outperforms state-of-the-art two-channel and multi-channel fusion approaches visually and quantitatively in common metrics, such as mutual information, visual information fidelity, and peak signal-to-noise ratio. The proposed model does not require manually tuned parameters, can be extended to an arbitrary number and arbitrary combinations of spectral channels, and is reconfigurable for addressing different use cases. We demonstrate examples for search and rescue, wildfire detection, and wildlife observation.</span> <strong><br></strong></p>
Aerial Images_Part 3_Integrating Remote Sensing and Machine Learning for Developing Spatio-Temporal Model to Predict Aquatic Larval Habitats of Malaria
<p>Aerial Images_Part 3_Integrating Remote Sensing and Machine Learning for Developing Spatio-Temporal Model to Predict Aquatic Larval Habitats of Malaria</p>
Annotated Cows in Aerial Images - Test Set for Livestock Detection Models
<p>This dataset contains a test set of aerial images from fields in Juchowo, Poland and Wageningen, the Netherlands, with annotated cows. The annotations are provided in a CSV format containing image file paths, bounding box coordinates (<code>xmin</code>, <code>ymin</code>, <code>xmax</code>, <code>ymax</code>), and labels for the objects (cows). The dataset is intended for evaluating livestock detection models in deep learning, particularly DeepForest.</p> <p>The images and their corresponding bounding box annotations are included in this archive. The test set consists of 10% of the total images, split from the original dataset created by G.J. Franke and Sander Mucher, which is available in [Harvard Dataverse](https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/N7GJYU). This dataset is part of the GenTORE project (<a href="https://www.gentore.eu" target="_new" rel="noopener">https://www.gentore.eu</a>), focusing on precision livestock farming using automated detection and deep learning techniques.</p> <p><strong>Contributions</strong></p> <p><em>Notebook</em></p> <ul> <li> <p>Alejandro Coca-Castro (author), The Alan Turing Institute, <a href="https://github.com/acocac">@acocac</a></p> </li> </ul> <p><em>Dataset reference and documentation</em></p> <ul> <li> <p>Franke, G.J., & Mucher, S. (2021). Annotated cows in aerial images for use in deep learning models. Harvard Dataverse. <a href="https://doi.org/10.7910/DVN/N7GJYU" target="_new" rel="noopener">https://doi.org/10.7910/DVN/N7GJYU</a></p> </li> </ul>
Raw images, annotations, and vvipr code archive to support 'Evaluating thermal and color sensors for automating detection of penguins and pinnipeds in images collected with an unoccupied aerial system''
<p>Images, annotations, and code archived here were used in the paper "Evaluating a machine learning approach to detect penguins and pinnipeds in thermal and color images collected with an unoccupied aerial system" submitted for publication in Drones. The files contain raw thermal and color images of aggregations of gentoo (<em>Pygoscelis papua</em>) and chinstrap (<em>P. antarcticus</em>) penguins and Antarctic fur seals (<em>Arctocephalus gazella</em>). All images were collected with the Flir DuoPro R camera (Teledyne FLIR LLC, Wilsonville, OR, U.S.A.), carried into flight under an APH-28 hexacopter (Aerial Imaging Solutions, LLC, Old Lyme, CT, U<strong>.</strong>S<strong>.</strong>A<strong>.)</strong> at Cape Shirreff, Livingston Island, Antarctica (60.79 °W, 62.46 °S), during the austral summer of 2019-20. All aerial surveys occurred under the Marine Mammal Protection Act Permit No. 20599 granted by the Office of Protected Resources/National Marine Fisheries Service, the Antarctic Conservation Act Permit No. 2017-012, NMFS-SWFSC Institutional Animal Care and Use Committee Permit No. SWPI 2014-03R, and all domestic and international UAS flight regulations. The annotations of the images were conducted using VIAME desktop software (v 0.16.1 or later;<a href="https://github.com/VIAME/VIAME">https://github.com/VIAME</a>) or the online using the DIVE interface (https://viame.kitware.com/). Model results were assessed with the vvipr code (v.0.3.2), archived here and available online (https://github.com/us-amlr/vvipr/releases/tag/v0.3.2).</p> <p> </p>
Dataset with square plots across Sierra Nevada (Spain) where the contours of all juniper shrubs were annotated as polygons using centimetric GPS and very high resolution aerial and satellite RGB images
<p><strong>This dataset is a shapefile of 767 polygons describing the contours of Juniperus communis L. and Juniperus sabina L. shrubs for the year 2021 in rectangular plots across Sierra Nevada. The coordinates of the polygons were obtained from a field work campaign with a differential centimetric GPS, and their contours were drawn manually in QGIS using the Google Earth satellite image for 2020 and the PNOA aerial image for the 2020. </strong></p> <p><strong>This dataset also contains an excel file describing the features of each polygon: the polygon centroid coordinates, the type of species, the sexgender, the morphotype, the damage in the vegetation cover estimated in the field and telematically, certainty of the digitalization with QGIS and also if the differential centimetric GPS used belongs to the University of Granada or the University of Almeria. </strong></p>
Unmanned Aerial Vehicle (UAV) image dataset.
<p>The dataset contains 2,919 images and separated into five classes of car, taxi, truck, bus and motorcycle.<br> </p>
Aerial RGB and Thermal Infrared (TIR) Images of Vineyards and Pseudo-coloring RGB Images of the Plant's Stressed Areas.
<p>This dataset consists of 375 high-resolution visible-spectrum (RGB) and 375 thermal infrared (TIR) images of a vineyard (Vitis vinifera L.) captured by a Unmanned Aerial Vehicle (UAV) carrying TIR and RGB sensors. Also, the dataset contains 375 RGB images with pseudo-coloring where plants' stressed areas exist, aligned, and cropped based on the TIR images' Field of View (FOV).</p>
A global data set of realized treelines sampled from Google Earth aerial images
<div> <span>We </span><span>sampled</span><span> Google Earth aerial images</span><span> to get a representative and globally distributed dataset of treeline locations</span><span>. </span><span>Google Earth images</span><span> are available to everyone, but may not be automatically downloaded and processed according to Google's license terms. Since we only wanted to detect tree individuals, we evaluated the aerial images manually by hand.</span> </div> <div> </div> <div> <span>Doing so, we scaled Google Earth's GUI interface to a buffer size of approximately 6000 m from a perspective of 100 m (+/- 20 m) above Earth's surface. Within this buffer zone, we took coordinates and elevation of the highest </span><span>realized </span><span>treeline locations. In some remote areas of Russia and Canada, individual trees were not identifiable due to insufficient image resolution. If this was the case, no treeline was sampled, unless we detected another visible treeline within the 6,000 m buffer and took this next highest treeline</span><span>. We did not ap</span><span>p</span><span>ly an automated image processing approach. </span><span>We calculated mass elevation effect as the distance to the nearest mountain chain limits. Continentality was assessed by the distance to the nearest coastline. Isolation was calculated by the nearest distance of a mountain chain to another mountain chain within a comparable elevational band. </span> </div>
Our processed LoveDA dataset for "LOANet: A Lightweight Network Using Object Attention for Extracting Buildings and Roads from UAV Aerial Remote Sensing Images"
<p>Our processed LoveDA dataset is used for the paper "<a href="https://doi.org/10.7717/peerj-cs.1467">LOANet: A Lightweight Network Using Object Attention for Extracting Buildings and Roads from UAV Aerial Remote Sensing Images</a>"</p>
Our processed CITY_OSM dataset for "LOANet: A Lightweight Network Using Object Attention for Extracting Buildings and Roads from UAV Aerial Remote Sensing Images"
<p>Our processed CITY_OSM dataset is used for the paper "<a href="https://doi.org/10.7717/peerj-cs.1467">LOANet: A Lightweight Network Using Object Attention for Extracting Buildings and Roads from UAV Aerial Remote Sensing Images</a>".</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.