Find research datasets worth reusing

opencc-by-4.0Mar 2020View details →

Drone aerial imagery of a small island in Kobbefjord (SW Greenland) acquired during Mission Arctic 2017

<code>Aerial images of a small island in Kobbefjord (SW Greenland) were acquired on 5 July 2017 using a DJI Phantom 3 Standard drone during the <a href="https://www.frontiersin.org/articles/10.3389/fmars.2021.665582/full">Mission Arctic citizen science expedition</a>. Images were processed in Agisoft Metashape. A digital elevation model (DEM) and an orthomosaic were exported at 5 cm resolution. For details, see the readme and processing report that accompanies this dataset. </code>

opencc-by-4.0Sep 2022View details →

Drone aerial imagery of a headland in Nuup Kangerlua (Godthåbsfjord, SW Greenland) acquired during Mission Arctic 2017

Aerial images of a headland in Nuup Kangerlua (Godthåbsfjord, Greenland) were acquired on 11 July 2017 using a DJI Phantom 3 Standard drone during the <a href="https://www.frontiersin.org/articles/10.3389/fmars.2021.665582/full">Mission Arctic citizen science expedition</a>. Images were processed in Agisoft Metashape. A digital elevation model (DEM) and an orthomosaic were exported at 2.5 cm resolution. A 5 cm resolution orthomosaic is also included. For details, see the readme and processing report that accompanies this dataset.

opencc-by-4.0Jul 2022View details →

An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery: Dataset

This archive contains code and data to go with the paper *An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery*.   This archive contains geospatial data, as well as the code used to generate the geospatial data. The geospatial data consists of georeferenced polygons identifying areas which are covered by green roofs in London (GBR) generated from 2019 aerial imagery. The data is described in detail in the manuscript *An Open-Source Automatic Survey of Green Roofs in London using Segmentation of Aerial Imagery*. See abstract below.   GeoJSON format: GeoJSON is a format for encoding geospatial data, see https://geojson.org/. GeoJSON can be read using GIS programs including ArcGIS, QGIS, OGR.   Contents: `geospatial_data/buffered_polygons_2021.zip` a zip archive containing a geojson file. It is the estimated locations of green roofs in London in 2021 and is the main result, which can be opened in any GIS program after being unzipped. `geospatial_data/buffered_polygons_2019.zip` a zip archive containing a geojson file. It is the estimated locations of green roofs in London in 2019 and is a secondary result, which can be opened in any GIS program after being unzipped. The predictions were made with the same model as the 2021 results. `geospatial_data/labelled_area.zip` a zip archive containing a geojson file. Identifies the area which was hand-labelled. `geospatial_data/manual_2021.zip` a zip archive containing a geojson file. Manually labelled green roof from 2021 imagery. `geospatial_data/manual_2019.zip` a zip archive containing a geojson file. Manually labelled green roof from 2019 imagery. `segmentation_code` contains the code used to produce the segmentation from the aerial imagery. `analysis_code` contains the code used to produce the plots and tables for the paper.   Imagery availability: Unfortunately the aerial imagery and building footprint data cannot be shared directly, as you will require the proper license. Both can be found at [Digimap](https://digimap.edina.ac.uk) provided your institution has the license.   Abstract: Green roofs can mitigate heat, increase biodiversity, and attenuate storm water, giving some of the benefits of natural vegetation in an urban context where ground space is scarce. To guide the design of more sustainable and climate resilient buildings and neighbourhoods, there is a need to assess the existing status of green roof coverage and explore the potential for future implementation. Therefore, accurate information on the prevalence and characteristics of existing green roofs is needed, but this information is currently lacking. Segmentation algorithms have been used widely to identify buildings and land cover in aerial imagery. Using a machine-learning algorithm based on U-Net to segment aerial imagery, we surveyed the area and coverage of green roofs in London, producing a geospatial dataset \cite[]{simpson_charles_2022_6861929}. We estimate that there was 0.23 km^2 of green roof in the Central Activities Zone (CAZ) of London, (1.07 km^2) in Inner London, and (1.89 km^2) in Greater London in the year 2021. This corresponds to 2.0% of the total building footprint area in the CAZ, and 1.3% in Inner London. There is a relatively higher concentration of green roofs in the City of London, covering 3.9% of the total building footprint area. Test set accuracy was 0.99, with an f-score of 0.58. When tested against imagery and labels from a different year (2019), the model performed just as well as a model trained on the imagery and labels from that year, showing that the model generalised well between different imagery. We improve on previous studies by including more negative examples in the training data, and by requiring coincidence between vector building footprints and green roof patches. We experimented with different data augmentation methods, and found a small improvement in performance when applying random elastic deformations, colour shifts, gamma adjustments, and rotations to the imagery. The survey covers 1558 km^2 of Greater London, making this the largest open automatic survey of green roofs in any city. The geospatial dataset is at the single-building level, providing a higher level of detail over the larger area compared to what was already available. This dataset will enable future work exploring the potential of green roofs in London and on urban climate modelling.

opencc-by-4.0Jun 2023View details →

Doodleverse/Segmentation Gym SegFormer models for 2-class (wood, other) segmentation of RGB aerial orthomosaic imagery

Doodleverse/Segmentation Gym SegFormer models for 2-class (wood, other) segmentation of RGB aerial orthomosaic imagery This model release is part of the Doodleverse: https://github.com/Doodleverse 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: 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. 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 File descriptions 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. 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. 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 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` 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` Additionally,  1. BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU 2. sample_images.zip contains a few example input files, for model testing 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   References *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 **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.

opencc-by-4.0Jun 2023View details →

Doodleverse/Segmentation Gym Residual Unet models for 2-class (alluvial sediment, other) segmentation of RGB aerial orthomosaic imagery

Doodleverse/Segmentation Gym Residual Unet models for 2-class (alluvial sediment, other) segmentation of RGB aerial orthomosaic imagery This model release is part of the Doodleverse: https://github.com/Doodleverse 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: 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. 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 File descriptions 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. 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. 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 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` 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` Additionally,  1. BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU 2. sample_images.zip contains a few example input files, for model testing References *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

opencc-by-4.0Jul 2023View details →

Doodleverse/Segmentation Gym SegFormer models for 4-class (other, water, sediment, wood) segmentation of RGB aerial orthomosaic imagery

Doodleverse/Segmentation Gym SegFormer models for 4-class (other, water, sediment, wood) segmentation of RGB aerial orthomosaic imagery This model release is part of the Doodleverse: https://github.com/Doodleverse 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 water, wood, sediment, and other in orthomosaic imagery. Models have been created using Segmentation Gym* using a dataset of images published here: 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. Classes: {0=other, 1=water, 2=sediment, 3=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 File descriptions 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. 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. 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 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` 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` Additionally,  1. BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU 2. sample_images.zip contains a few example input files, for model testing References *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

opencc-by-4.0Jul 2023View details →

Doodleverse/Segmentation Gym SegFormer models for 2-class (other, sediment) segmentation of RGB aerial orthomosaic imagery

Doodleverse/Segmentation Gym SegFormer models for  2-class (other, sediment)  segmentation of RGB aerial orthomosaic imagery This model release is part of the Doodleverse: https://github.com/Doodleverse 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 water, wood, sediment, and other in orthomosaic imagery. Models have been created using Segmentation Gym* using a dataset of images published here: 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. 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 File descriptions 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. 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. 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 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` 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` Additionally,  1. BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU 2. sample_images.zip contains a few example input files, for model testing References *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

edi44/100

Kelp canopy chlorophyll to carbon ratio derived from aerial hyperspectral imagery

This dataset represents a time series of giant kelp canopy chlorophyll to carbon ratio (Chl:C) derived from aerial hyperspectral imagery, along with associated environmental, canopy age determinations, and validation datasets. Additional data from a frond cohort experiment are also presented, representing empirical observations of the decline in chlorophyll pigment concentration with blade age. This dataset contains eight geotiff rasters showing canopy Chl:C at a 30-meter pixel resolution for giant kelp forests in the Santa Barbara Channel during the months of April, June, and August from 2013 – 2015.

openCC (other)Aug 2021View details →

opencc-by-4.0Jul 2020View details →

Data and R code for the revised manuscript "Downscaling digital soil maps using electromagnetic induction and aerial imagery"

Data and R code for the revised manuscript "Downscaling digital soil maps using electromagnetic induction and aerial imagery". This is the code for the revised version of the manuscript, after adressing comments from reviewers. The data and code for the preprint, before submission to peer review (Møller et al., 2020), is available at <a href="https://doi.org/10.5281/zenodo.3699130">https://doi.org/10.5281/zenodo.3699130</a>. The R code was written for R version 3.6.3. References Møller, A.B., Koganti, T., Beucher, A., Iversen, B.V. and Greve, M.H., 2020. Downscaling digital soil maps using electromagnetic induction and aerial imagery. EarthArXiv. <a href="http://dx.doi.org/10.31223/osf.io/a7xz6">http://dx.doi.org/10.31223/osf.io/a7xz6</a>. [preprint]

opencc-by-4.0Sep 2020View details →

Drone aerial imagery of macroalgae growing in a small embayment at the western end of the Nuussuaq Peninsula in Northwest Greenland

Aerial images of a small embayment at the western tip of the Nuussuaq Peninsula (NW Greenland) were acquired on 10 August 2019 using a DJI Phantom 3 Standard quadcopter drone. The imagery was acquired during the Vaigat Iceberg-Microbial Oil Degradation and Archaeological Heritage Investiga- tion (VIMOA) research cruise. Overlapping images were acquired during two consecutive flights using the 'interval' setting with a period of 5 seconds in the DJI Go app. The drone was flown manually. This drone survey was carried out with two primary objectives in mind: 1) Identify and quantify the macroalgae growing on the coast; 2) Identify any archaeological and cultural heritage sites that may be present. Images were processed using Agisoft PhotoScan Pro (v1.4.4; Linux Ubuntu). Images were aligned using the script CARMA_PhotoScan_Align.py (see this script for settings). After image alignment, the bounding box was visually inspected and the sparse point cloud was thinned using thresholds for the reconstruction uncertainty, projection accuracy, and reprojection error (gradual selection).  The remaining tie points in the sparse cloud were then used to build the dense cloud. The settings used for this step can be found in the script CARMA_PhotoScan_GS_DC_v2.py. Next, the dense point cloud was used to generate a digital elevation model (DEM) and, subsequently, an orthomosaic. The Agisoft PhotoScan processing report is also included.

opencc-by-4.0Dec 2020View details →

Drone-based aerial imagery of rivers, wetlands and agricultural systems in Zambia

Unmanned aerial vehicle (UAV) imagery of rivers, wetlands and agricultural systems across Zambia. Collected during two flying seasons in March and September 2018, with a total of 122 scenes at 48 sites. All imagery is made available on OpenAerialMap (<a href="https://map.openaerialmap.org">https://map.openaerialmap.org</a>), a set of tools for searching, sharing, and using openly licensed satellite and UAV imagery. Detailed description of the dataset provided in a .csv file.

opencc-by-4.0Oct 2024View details →

MultiFranceFences: A novel deep learning dataset for automated fence detection from multimodal aerial imagery

The MultiFranceFences dataset is a large-scale, multimodal remote sensing benchmark for the semantic segmentation of fences across various landscapes in France. This dataset integrates high-resolution orthophotographs (RGB through BDOrtho) and Digital Surface Models (DSM) derived from LiDARHD data.  MultiFranceFences is suitable for deep learning models in semantic segmentation, including state-of-the-art models like UNet, D-LinkNet, and the newly proposed H-IncepUNet, which integrates handcrafted features and multi-scale feature extraction modules for enhanced fence detection. Dataset features: <ul> <li>Multimodal imagery: Combines orthophotographs and DSM data from LiDARHD for fences semantic segmentation (folders ortho and lidar).</li> <li>Buffer options: 2-meter and 3-meter buffer fence annotations to fit varying detection requirements (folders fences_2m and fences_3m).</li> <li>Diverse landscapes: Covers rural, and natural environments across France.</li> <li>Validated dataset: Manually cleaned and validated to remove erroneous fence labels under tree canopies or areas with limited visibility.</li> </ul> Each patch is named according to the nomenclature of the original BDOrtho tile, followed by the specific x and y coordinates of the patch within that tile.

opencc-by-4.0Aug 2021View details →

Multi-resolution dataset for photovoltaic panel segmentation from satellite and aerial imagery

A <a href="https://www.sciencedirect.com/topics/engineering/photovoltaics">photovoltaic</a> (PV) dataset from satellite and aerial imagery. The dataset includes three groups of PV samples collected at the spatial resolution of 0.8m, 0.3m and 0.1m, namely PV08 from Gaofen-2 and Beijing-2 imagery, PV03 from aerial photography, and PV01 from UAV orthophotos. PV08 contains rooftop and ground PV samples. Ground samples in PV03 are divided into five categories according to their background land use type: shrub land, grassland, cropland, saline-alkali, and water surface. Rooftop samples in PV01 are divided into three categories according to their background roof type: flat concrete, steel tile, and brick. Data document can refer to the preprint https://essd.copernicus.org/preprints/essd-2021-270/

opencc-by-4.0Aug 2021View details →

Paired Waikato Region Aerial Photography and Sentinel-2 Imagery

This dataset contains paired high resolution orthorectified aerial photography provided by the Waikato Region Aerial Photography initiative, paired with Sentinel-2 satellite images. These images were collected as part of a satellite imagery super-resolution research project. The aerial photograph was down-sampled to a spatial resolution of 2.5m per pixel, while the satellite images were taken at a spatial resolution of 10m per pixel. The satellite images were taken from 8 fly-bys between 16th January to 2nd March 2019 so that it is temporally consistent with the aerial photographs taken in February of 2019.

opencc-by-4.0Sep 2022View details →

Drone aerial imagery of a macroalgal-covered coastline in Kobbefjord (SW Greenland) in 2018 and 2019

Aerial images of inner Kobbefjord (SW Greenland) were acquired on two occasions in 2018 (2018-09-07 15:06; 2018-09-07 16:00) representing different tidal levels and on one occasion in 2019 (2019-08-26). Images were acquired using a DJI Mavic Air drone. The imagery was acquired in connection with the “Greenland Ecosystem Monitoring Program” “Nuuk Basis” monitoring of marine flora (http://<a href="http://g-e-m.dk/">g-e-m.dk</a>/). This drone survey was carried out with the primary goal of documenting and quantifying the area distribution of the tidal vegetation (dominated by the brown macroalga Ascophyllum nodosum) and shallow subtidal vegetation at the monitoring site. Images were processed using Agisoft Metashape Professional to produce high resolution orthomosaics from each UAV survey.

opencc-by-4.0Jun 2024View details →

Waikato Multi Label Aerial Imagery Dataset 2012

This dataset is a supplementary dataset to https://zenodo.org/records/11484867. The images in this dataset are spatially coincident, taken ~5 years prior to the referenced dataset.  Below is the description of the referenced dataset Effective land use management is crucial for balancing the development against environmental sustainability, preservation of biodiversity and resilience to climate change impacts. Despite this, there is a notable scarcity of comprehensive aerial imagery datasets for refining and improving machine learning frameworks for informed policy making. In this paper, we introduce a substantial aerial imagery dataset from New Zealand curated for the Waikato region spanning 25,000 km^2, specifically to address this gap and empower global research efforts. The dataset comprises of a main set, containing more than 140,000 images, with 3 supplementary sets. Each image is annotated with 33 fine-grained, multi-labelled classes and approximate segmentation masks of the classes, with 3 supplementary datasets covering spatially coincident satellite imagery and aerial imagery 5 years prior and 5 years later from the main dataset.

opencc-by-4.0Jun 2024View details →

Waikato Multi Label Aerial Imagery Dataset

Effective land use management is crucial for balancing development against environmental sustainability, preservation of biodiversity, and resilience to climate change impacts. Despite this, there is a notable scarcity of comprehensive aerial imagery datasets for refining and improving machine learning frameworks to better inform policy making. In this paper, we introduce a substantial aerial imagery dataset from New Zealand curated for the Waikato region, spanning 25,000 km\(^2\), specifically to address this gap and empower global research efforts. The dataset comprises of a main set, containing more than 140,000 images, with three supplementary sets. Each image in the main dataset is annotated with 33 fine-grained, multi-labeled classes and approximate segmentation masks of the classes, and the three supplementary datasets cover spatially coincident satellite imagery and aerial imagery five years prior and five years later from the main dataset.

opencc-by-4.0Jun 2024View details →

Waikato Multi Label Aerial Imagery Dataset 2023

This dataset is a supplementary dataset to https://zenodo.org/records/11484867. The images in this dataset are spatially coincident, taken ~5 years after the referenced dataset.  Below is the description of the referenced dataset Effective land use management is crucial for balancing the development against environmental sustainability, preservation of biodiversity and resilience to climate change impacts. Despite this, there is a notable scarcity of comprehensive aerial imagery datasets for refining and improving machine learning frameworks for informed policy making. In this paper, we introduce a substantial aerial imagery dataset from New Zealand curated for the Waikato region spanning 25,000 km^2, specifically to address this gap and empower global research efforts. The dataset comprises of a main set, containing more than 140,000 images, with 3 supplementary sets. Each image is annotated with 33 fine-grained, multi-labelled classes and approximate segmentation masks of the classes, with 3 supplementary datasets covering spatially coincident satellite imagery and aerial imagery 5 years prior and 5 years later from the main dataset.