Hall, Emma; Lara, Mark J (2024): UAS-multisensor input imagery and random forest classification products for mapping vegetation species at a tallgrass prairie in Urbana, IL [dataset]. PANGAEA, https://doi.org/10.1594/PANGAEA.967351
Always quote citation above when using data! You can download the citation in several formats below.
Published: 2024-05-14 • DOI registered: 2024-05-14
Abstract:
Three mosaic raster files are provided depicting Uncrewed Aerial Systems (UAS) data collection from RGB, multispectral, and hyperspectral sensors, which were collected at Weaver Park in Urbana, IL. Along with spectral bands, all image products are fused with canopy height models (CHM), where RGB and multispectral products are fused with an SfM-derived (structure from motion) CHM and the hyperspectral product is fused with a LiDAR-derived (Light Detection and Ranging) CHM. A multispectral phenological time series product is also provided, where Normalized Difference Vegetation Index (NDVI) was calculated across six time periods in one growing season and where an additional 15 NDVI-derived metrics were calculated, resulting in a 21-band image product. Differential GPS (dGPS; 2cm resolution) data identifying vegetation species are also provided, which were used at the training and testing datasets for conducting random forest classifications on each image product. Random forest classification models were applied to each of the three UAS-sensor types both with and without CHM fusion, and to the multispectral phenology time series image, resulting in the 7 vegetation maps provided.
Keyword(s):
Related to:
Hall, Emma; Lara, Mark J (2022): Multisensor UAS mapping of Plant Species and Plant Functional Types in Midwestern Grasslands. Remote Sensing, 14(14), 3453, https://doi.org/10.3390/rs14143453
Funding:
National Science Foundation (NSF), grant/award no. 1928048: Collaborative Research: Navigating Disturbance Regimes in the New Arctic
Coverage:
Median Latitude: 40.110329 * Median Longitude: -88.180550 * South-bound Latitude: 40.109810 * West-bound Longitude: -88.181343 * North-bound Latitude: 40.110847 * East-bound Longitude: -88.179757
Date/Time Start: 2020-05-01T00:00:00 * Date/Time End: 2020-10-31T23:59:59
Event(s):
Comment:
All data files have WGS 1984 World Mercator coordinate system
Parameter(s):
| # | Name | Short Name | Unit | Principal Investigator | Method/Device | Comment |
|---|---|---|---|---|---|---|
| 1 | File content | Content | Hall, Emma | |||
| 2 | Binary Object | Binary | Hall, Emma | |||
| 3 | Binary Object (File Size) | Binary (Size) | Bytes | Hall, Emma |
License:
Creative Commons Attribution 4.0 International (CC-BY-4.0)
Status:
Curation Level: Basic curation (CurationLevelB)
Size:
12 data points
Data
All files referred to in data matrix can be downloaded in one go as ZIP or TAR. Be careful: This download can be very large! To protect our systems from misuse, we require to sign up for an user account before downloading.
| 1 Content | 2 Binary | 3 Binary (Size) [Bytes] |
|---|---|---|
| 11-band product where first 10 bands are from MNF rotation of UAS-hyperspectral imagery and Band 11 is the LiDAR-derived CH | hyperspectral_CHM_img.7z | 782.8 MBytes |
| 7-band product where bands 2-6 are UAS-multispectral imagery product and band 1 is SfM-derived CHM. Band 7 is noise | multispectral_CHM_img.7z | 606.9 MBytes |
| 22-band product where band 22 is NoData and bands 1 and 17-21 are the NDVI of six multispectral time series images (Figure 3 in publication) and bands 2-16 are phenology metrics defined in Table 3 of publication | multi_NDVImetrics_img.7z | 2.1 GBytes |
| 5-band product where bands 1-3 are RGB, band 4 is noise, and band 5 is SfM-derived CHM | RGB_CHM_img.7z | 239.2 MBytes |
| Input dGPS point dataset used in Random Forest Classifications | Input_dGPS_vegData_shp.zip | 21.9 kBytes |
| Random Forest classification vegetation maps for each sensor product with and without CHM added | RF_Output_Files.7z | 99.8 MBytes |