Process_Description:
GROUND CONTROL: Five AeroPoint GCPs were spaced out over the field site and left on for at least two hours to collect GNSS data. After collection and connected to Wifi, the AeroPoint data were uploaded and run through a post-processing kinematic algorithm of the CORS network to get a global accuracy. Internal accuracy is reported in xyz variance and with a baseline distance. The data were exported in the horizontal datum NAD83(2011) to produce latitude, longitude, and ellipsoid heights, and then UTM19N and vertical datum NAVD88 using Geoid 18 was used to produce easting and northing and orthometric heights. These were exported to a CSV file and named 2024004FA_MI_Aug_AeroPoints.csv. A SP80 base station was set up on a brass disk to record GNSS data for post processing the lidar data. An Emlid RS3 with tilt compensation turned on was connected to MASS CORS was used to take five-second averaged check shots of the creek center, data is provided in 2024004FA_MI_Aug_Emlid_RS3.csv
Process_Date: 20240821
Process_Description:
UAS FLIGHTS: The lidar sensor is a 905 nm wavelength Livox Avia with a 70.4/ 4.5 degree horizontal and vertical field of view (FOV). The Lidar scanner was set to the non-repetitive scan pattern, which moves the sensor in spirograph (flower) motion and is optimal for vegetated areas as it can capture more angles by surveying in front and behind the sensor. The camera module is a SONY UMC - 10RC, a 20.4 megapixel (MP). The system also consists of a Trimble AV18 GNSS antenna mounted to the top of the UAS and connected to the lidar system via GNSS cable. The lidar data is saved to a 256 GB USB thumb drive in three different files: (1) the IMUPGPS data, decimated for quick post-processing, in binary *.ys format, (2) the scanner data in *.lvx format (~600 MB per minute of data collection), (3) and the complete IMUPGPS data in Applanix (Trimble) binary *.t04 format. The RGB images are saved to a 64 GB micro SD card as *.jpeg files. A configuration text file (CONFIG.TXT) is pre-loaded onto the USB thumb drive and can be edited to control lidar and camera module settings, including the camera triggering height, the camera triggering interval, and the lidar scan pattern.
The aPT has 7 sensors with global shutters: 5 multispectral bands, blue (443-507 nm), green (533-587 nm), red (652-684 nm), red edge (705-729 nm), and near-infrared (NIR) (785-899 nm), 1 panchromatic band (171.5 nm - 1098 nm), and 1 long-wave infrared (LWIR) thermal band (4.5 - 16.5 microns). The LWIR band images were not used in any processing workflows and were excluded from this data release. The multispectral bands have a focal length of 8 mm and acquires images at 3.2 megapixels, 2064 x 1544 pixel resolution, and a 12-bit depth. The panchromatic band has a focal length of 16.3 mm and acquires images at 12 megapixels, 4112 x 3008 pixel resolution, and 12-bit depth. The sensor apertures are f/1.8 for the spectral bands and f/4.5 for the panchromatic band. Sensor properties are automatically written to the image exif. Images are saved to a CF Express type B SD card in TIFF format. *Note that because most 3rd-party software supports 16-bit images, the 12-bit images are output to TIFF files in a 16-bit container with the last four bits filled with zeros.
Both sensors (YSMP and aPT) were attached to an DJI Matrice 600 Pro UAS with approved government edition firmware. Two 1x1 m Group 8 Technology reflectance calibration tarps with reflectance values of 0.48 and 0.12 were staked out within the UAS survey area for validation of post-processed aPT reflectance products.
The YSMP lidar data was collected with the UAS flying at 10 m/s at 90 meters above ground level with north-south and east-west transect passes. The scan method was set to repetitive. The camera module was set to take images every 2 seconds. After the flight the lidar data were taken off the sensor.
The aPT data was collected with the UAS flying at 8 m/s at 60 meters above ground level with north-south and east-west transect passes that achieve ~80% forelap and sidelap. Image sets were taken every 2 seconds. The aPT received power from a connection to the UAS. The downwelling light sensor was attached and measured ambient light for each of the spectral bands captured by the camera. For each camera shutter trigger, camera settings (focal length, ISO, etc), spectral irradiance, roll, pitch, yaw, gain, exposure time, and GPS position data were automatically incorporated into the Exif data for each image. Just prior to takeoff and after landing, a Micasense calibration panel (RP06-2224002-OB) was placed under the camera for several seconds to capture calibration images. After the flight, the UAS was powered off, the SD card was removed, and all images and files relating to the survey were downloaded to a field computer.
The Skydio X10 is a small UAS with an integrated VT300-L camera system. The radiometric thermal camera is a FLIR Boson+ with less than 30 millikelvin sensitivity. The Skydio X10 also has two RGB cameras, one has a 50 degree field of view (narrow) and the other has a 93 degree field of view (wide). A single flight in auto mapping mode was flown at 60 meters with 70% desired sidelap and overlap and 1.21 cm ground sampling distance. The UAS was flown at 5 m/s using the wide camera and thermal camera. Two sets of thermal images are taken, but only the set with the radiometric data (SX10r) is provided, as the second set is redundant.
Note, the aPT and SX10 geotagged positions embedded in the imagery exif information are in WGS84 and Geoid EGM96 and for the YSMP photos it is WGS84 and ellipsoid height, this is how the data are collected. The positions have been converted to NAD83(2011) and NAVD88 geoid 18 in the imagery locations file (2024004FA_MI_Aug_ImageryLocations.csv) and is accounted for when transforming to NAD83(2011)/UTM19N and NAVD88 Geoid 18 in the products.
Process_Date: 20240821
Process_Description:
RAW IMAGERY: The aPT and SX10 images are automatically geotagged from the UAS GPS, the YSMP images were geotagged in Emlid Studio using the lidar .T04 file and base station rinex files. All images were processed to add additional information required by the USGS to the EXIF headers using ExifTools (
https://exiftool.org/, version: 12.06), and the files were renamed to a unique identifier using Namexif (
http://www.digicamsoft.com/softnamexif.html, version 2.1) to avoid any possibility of duplicate names. These steps are described here. Note that in the commands below .*JPG was used for SX10 and YSMP images and *.tif was used for the aPT images)
1. ExifTools was used to tag each photo headers following the Imagery Data System EXIF Guidance. Attributes (e.g. Credit, Copyright, UsageTerms, ImageDescription, Artist, etc) were stored in a csv file and written to each image with the command:' exiftool -csv="C:\directory\name\EXIF.csv" C:\directory\name\of\photos *.JPG '
To read out the photo information to a csv when in the directory with the photos the command is: exiftool -csv *.JPG > directory/name/allheaders_out.csv
2. All the images were renamed with Namexif (
https://us.digicamsoft.com/softnamexif.html v 2.2 accessed April 2020) to ensure unique filenames and compliance with the USGS Coastal and Marine Hazards and Resources Program's best practices for image naming convention. Images were renamed with the field survey ID prefix; flight number, and ID that distinguishes USGS cameras by make/camera number, the image acquisition date, coordinated universal time (UTC) in ISO8601 format, and a suffix with the original image name. For example, image name '2024004FA_f01YSMP_20231026T165822Z_IMG_####_#', 2024004FA is the field activity ID, f01 is the flight number, YSMP is the camera on the YellowScan Mapper Plus and aPT is the altum-PT, 20231026 is the UTC date in the format YYYYMMDD, and a 'T' is used to separate UTC date from UTC time in format HHMMSS followed by a Z. The IMG_####_# is the original raw photo name appended to the end of the new filename; for the aPT there are images with the same time and photo name distinguished by the band number _#.
3. Images are validated and uploaded onto the Imagery Data System based on sensor and image type.
Process_Description:
PHOTOGRAMMETRY: The three ortho products were created in Agisoft Metashape v. 2.0.1 using the following general steps (see Over and others, 2021 for a more detailed SfM methodology explanation,
https://support.micasense.com/hc/en-us/articles/4416696717847-Pan-sharpening-processing-data-from-Altum-PT-and-RedEdge-P-cameras-in-Agisoft-Metashape for processing the aPT images, and
https://support.skydio.com/hc/en-us/articles/27236378437019-Scan-reconstruction-best-practices-with-Skydio-X10 for processing the SX10 images):
1. For each image type (SX10r, SX10c, aPT) a project was created and imagery was imported (as a multi-camera system for SX10r and the aPT images).
2. Photos were aligned at a low accuracy and then GCPs were automatically detected in the point cloud. For the aPT images, the Panchromatic band images were used to identify tie points. GCP positions (2024004FA_MI_Aug_AeroPoints.csv) were added to the project in the reference systems NAD83(2011)/UTM Zone 19N and NAVD88 (geoid 18). The horizontal and vertical accuracies for the GCPs were set to 0.01/0.02 m, respectively and the camera positions for the images were turned off. The photos were then re-aligned with high accuracy (the pixels were not subsampled) using a keypoint limit of 60,000 and unlimited tie points.
3. The alignment process matched pixels between images to create point clouds and put the imagery into a relative spatial context using the GCPs. The resultant point clouds were filtered using one iteration of the 'Reconstruction uncertainty' filter at a level of 13, one iteration of the 'Projection accuracy' filter at a level of 3, and three iterations of the 'Reprojection accuracy' filter to get to a level of 1. With each filter, iteration points are selected, deleted, and then the camera model was optimized to refine the focal length, cx, cy, k1, k2, k3, p1, and p2 camera model coefficients.
4. At this point, for each project multiple ‘chunks’ were created so that independent high quality dense clouds with a low-frequency filtering algorithm could be made. The dense point clouds were then edited by visual inspection to remove points with a low confidence near the edges and near water bodies.
5. A DSM is built from the dense point cloud and then an orthomosaic is built from the DSM with refined seamlines. The SX10c orthomosaic is a 3-band RGB-averaged orthomosaic exported at 5 cm resolution (2024004FA_MI_Aug_RGBavg_Orhto_5cm.tif). The SX10r orthomosaic is single-band (IR), the values are converted from Kelvin to Celsius using the band math (0.01*thermal_band)-273.15 and exported at 5 cm resolution keeping only values within the ranges of 11 and 40 to remove outlier values (2024004FA_MI_Aug_SX10_IR_Ortho_5cm.tif). The aPT ortho product was pansharpened and transformed so that the bands were all divided by the reflectance normalization factor 32768 (half of the available digital number range or 100% reflectance for each band for 16-bit images). The normalization process produces values that are a unitless reflectance value between 0 and 1, where 1 would be 100% reflectance, i.e. lighter. This 5-band product was exported at 5 cm resolution (2024004FA_MI_Aug_aPT_5Band_Ortho_5cm.tif).
Process_Description:
LiDAR DATA: The YSMP lidar data were processed in Yellowscan CloudStation software integrated with Trimble POSPac UAV 9.0. Raw scanner data (.ys file) was imported into YellowScan CloudStation, the sensor LiDAR (.profile) and Camera (.camera) profiles, provided by the vendor, were selected for the project and the project coordinate reference system is set. CloudStation flight trajectories are adjusted manually to select the desired data to be processed. The .T04 file and base station GNSS rinex file were used to correct and optimize the sensor position trajectories and produce a Smoothed Best Estimate of Trajectory (SBET) file in .txt ASCII format, which represents the Post Processing Kinematic (PPK) Solution. The lever arm offsets and boresight angle corrections were applied to the LPC, along with a strip adjustment between transect swaths using CloudStation's "robust" setting. The LPC was then classified in CloudStation using a bare-ground classification scheme (0-Created, never classified; 2-Ground) with the settings Object inner size = 55 m, Steepness = 0.05 m, minimal object height = 0.50 m, and a point cloud thickness of 0.15 m. The LPC was colored by the YSMP camera based on the photo timestamp. The classified LPC is used to export a DTM of the mean values of just the classified ground points (2024004FA_MI_Aug_YSMP_DTM_5cm.tif). The classification is then cleared in Global Mapper (v 26.0) so all points are unclassified. Points are also deleted in Global Mapper that are interpreted to be noise or are above a reasonable elevation threshold based on the surrounding features. The clean LPC is exported as a zipped laz file (v 1.4) (2024004FA_MI_Aug_YSMP_LPC.laz). A DSM using maximum values of all points is exported (2024004FA_MI_Aug_YSMP_DSM_5cm.tif). All files were exported in EPSG:6348 (NAD83(2011)/UTM zone 19N and the vertical datum in EPSG:5703 (NAVD88 height in meters).
Process_Date: 20240902
