Process imagery taken during the SnowEx Alaska 2022 and 2023 fall campaign to reconstruct the snow pit ground surface roughness. The used ground control markers when capturing the images can be found under the docs folder.
The initial imagery recording and processing is inspred by Meloche et al. (2021) with updates to recording setup.
Central entry point to pass to Agisoft Metashape is the process-images.py script. A sample call can be found under the RCS folder.
Linux example usage:
metashape.sh -platform offscreen -r process-images __options__
required options:
-pn/--project-name
-op/--output-path
-if/--image-folder
-mf/--marker-file
all options:
-h, --help show this help message and exit
-pn PROJECT_NAME, --project-name PROJECT_NAME
Name of project.
-op OUTPUT_PATH, --output-path OUTPUT_PATH
Output directory for the Metashape project.
-if IMAGE_FOLDER, --image-folder IMAGE_FOLDER
Location of images relative to base-path.
-it IMAGE_TYPE, --image-type IMAGE_TYPE
Type of images - default to .jpg
-mf MARKER_FILE, --marker-file MARKER_FILE
Path to CSV file with marker distances
-dcq {1,2,4}, --dense-cloud-quality {1,2,4}
Integer for dense point cloud quality.
Highest -> 1 (Default) High -> 2 Medium -> 4
-exp, --export Export the PDF report and LAZ point cloud
--export-only Only run the export for the PDF report and LAZ point cloud.
NO processing will be performed.
See the entwine instructions on how to visualize points clouds with a cloud optimized data format on your local computer.
Below the instructions to develop the Metashape automation locally. This does not require an installed version with a license of Metashape.
- Setup conda environment
conda create -n gsr2 python
- Download the Metashape Python package from their website
pip install /path/to/downloaded/metashape.whl
- Install into the conda environment
conda activate gsr2 pip install /path/to/metashape.whl
J. Meloche , A. Royer , A. Langlois , N. Rutter & V. Sasseville (2020): Improvement of microwave emissivity parameterization of frozen Arctic soils using roughness measurements derived from photogrammetry, International Journal of Digital Earth, DOI: 10.1080/17538947.2020.1836049