UAV Remote Sensing Imagery of Miscanthus — 2020
CABBI Theme: Feedstock Production
Keywords: Biomass Analytics, Remote Sensing
Varela Quintela, S., Leakey, A.D.B., Sacks, E. October 10, 2022. “UAV remote sensing imagery — Miscanthus trials 2020 — Energy Farm — UIUC.” University of Illinois Urbana-Champaign. DOI: 10.13012/B2IDB-5689586_V1.
Aerial imagery utilized as input in the manuscript “Deep Convolutional Neural Networks Exploit High-Spatial- and -Temporal-Resolution Aerial Imagery to Predict Key Traits in Miscanthus.” Data was collected over Miscanthus sacchariflorus (Msa) and Miscanthus sinensis (Msi) breeding trials at the University of Illinois Energy Farm in 2020. Flights were performed using a DJI M600 mounted with a Micasense Rededge multispectral sensor at 20 m altitude around solar noon. Imagery is available as tif file by field trial and date (10). The post-processing of raw images into orthophoto was performed in Agisoft Metashape software. Each crop surface model and multispectral orthophoto was stacked into a unique raster stack by date and uploaded here. Each raster stack includes 6 layers in the following order: Layer 1 = crop surface model, Layer 2 = Blue, Layer 3 = Green, Layer 4 = Red, Layer 5 = Rededge, and Layer 6 = NIR multispectral bands. Msa raster stacks were resampled to 1.67 cm spatial resolution and Msi raster stacks were resampled to 1.41 cm spatial resolution to ease their integration into further analysis. ‘MMDDYYYY’ is the date of data collection, ‘MSA’ is M. Sacchariflorus trial, ‘MSI’ is Miscanthus Sinensis trial, ‘CSM’ is crop surface model layer, and ‘MULTSP’ are the five multispectral bands.
Data Download (3.5 KB) includes:
- Data collection timing
- Approach description
- CNN model description
Varela, S., Zheng, X., Njuguna, J.N., Sacks, E.J., Allen, D.P., Ruhter, J., Leakey, A.D.B. Oct. 25, 2022. “Deep Convolutional Neural Networks Exploit High-Spatial- and -Temporal-Resolution Aerial Imagery to Phenotype Key Traits in Miscanthus.” Remote Sensing 2022, 14(21), 5333. DOI: 10.3390/rs14215333.