satproc

Python library and CLI tools for processing geospatial imagery for ML

Description

satproc helps you work with large amount of geospatial raster images (satellite, drone, etc.) and process them for training machine learning, for object detection or semantic segmentation problems.

Installation

To install the latest stable version run:

pip install pysatproc

You can also clone the repository at https://github.com/dymaxionlabs/satproc/ and install the development version with the -e option, like this:

git clone https://github.com/dymaxionlabs/satproc.git
pip install -e satproc/

Now, whenever you want to bring the latest changes, just run git pull from the cloned repository.

Usage

Command Line (CLI)

When installed, satproc makes available a series of command-line scripts to process files without resorting to writing a Python script.

• satproc_extract_chips: Extract chips from raster images, optionally creating masks for each chip using a labels vector file.
• satproc_make_masks: Create masks from raster images and a labels vector file.
• satproc_polygonize: Polygonizes chip images into a single polygon vector file.
• satproc_generalize: Generalizes vector files by simplyfing and smoothing polygon boundary lines.
• satproc_smooth_stitch: Smoothes overlapping probability result chips.
• satproc_scale: Rescales values from raster images
• satproc_match_histograms: Matches histograms of raster images from a reference image.

Run any command with the -h/--help flag to see the available options and information on how to use them.

Contributing

Bug reports and pull requests are welcome on GitHub at the issues page. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant code of conduct.

License

This project is licensed under Apache 2.0. Refer to LICENSE.txt.