i.sam2 - GRASS GIS manual

Note: This document is for an older version of GRASS GIS that has been discontinued. You should upgrade, and read the current manual page.

NAME

i.sam2 - Integrates SAMGeo model with text prompt for segmentation in GRASS GIS.

KEYWORDS

imagery, segmentation, object recognition, deep learning

SYNOPSIS

i.sam2

i.sam2 --help

i.sam2 group=name output=name [checkpoint_dir=name] [text_prompt=string] [text_threshold=float] [box_threshold=float] [--overwrite] [--help] [--verbose] [--quiet] [--ui]

Flags:

--overwrite: Allow output files to overwrite existing files
--help: Print usage summary
--verbose: Verbose module output
--quiet: Quiet module output
--ui: Force launching GUI dialog

Parameters:

group=name [required]: Name of input imagery group
output=name [required]: Name of output segmented raster map
checkpoint_dir=name: Path to the SAMGeo model checkpoint directory (optional if using default model)
text_prompt=string: Text prompt to guide segmentation
text_threshold=float: Text threshold for text segmentation; Default: 0.24
box_threshold=float: Box threshold for text segmentation; Default: 0.24

DESCRIPTION
REQUIREMENTS
EXAMPLES
NOTES
REFERENCES
SEE ALSO
AUTHOR

DESCRIPTION

i.sam2 allows users to segment orthoimagery based on text prompts using SamGeo.

REQUIREMENTS

        pip install pillow numpy torch segment-geospatial

EXAMPLES

Segment orthoimagery using SamGeo2:

    i.sam2 group=rgb_255 output=tree_mask text_prompt="trees"

NOTES

The first time use will be longer as the model needs to be downloaded. Subsequent runs will be faster. Additionally, CUDA is required for GPU acceleration. If you do not have a GPU, you can use the CPU by setting the environment variable `CUDA_VISIBLE_DEVICES` to `-1`.

REFERENCES

Wu, Q., & Osco, L. (2023). samgeo: A Python package for segmenting geospatial data with the Segment Anything Model (SAM). Journal of Open Source Software, 8(89), 5663. https://doi.org/10.21105/joss.05663
Osco, L. P., Wu, Q., de Lemos, E. L., Gonçalves, W. N., Ramos, A. P. M., Li, J., & Junior, J. M. (2023). The Segment Anything Model (SAM) for remote sensing applications: From zero to one shot. International Journal of Applied Earth Observation and Geoinformation, 124, 103540. https://doi.org/10.1016/j.jag.2023.103540

AUTHOR

Corey T. White (NCSU GeoForAll Lab & OpenPlains Inc.)

SOURCE CODE

Available at: i.sam2 source code (history)

Latest change: Friday Feb 21 16:56:09 2025 in commit: e9df778894099cc01bf63c325fa7b86db4e6c128