GeodisTK: Geodesic Distance Transform Toolkit for 2D and 3D Images
Geodesic transformation of images can be implementated with two approaches: fast marching and raster scan. Fast marching is based on the iterative propagation of a pixel front with velocity F [1]. Raster scan is based on kernel operations that are sequentially applied over the image in multiple passes [2][3]. In GeoS [4], the authors proposed to use a 3x3 kernel for forward and backward passes for efficient geodesic distance transform, which was used for image segmentation.
Raster scan for geodesic distance transform. Image from [4].
DeepIGeoS [5] proposed to combine geodesic distance transforms with convolutional neural networks for efficient interactive segmentation of 2D and 3D images.
- [1] Sethian, James A. "Fast marching methods." SIAM review 41, no. 2 (1999): 199-235.
- [2] Borgefors, Gunilla. "Distance transformations in digital images." CVPR, 1986
- [3] Toivanen, Pekka J. "New geodesic distance transforms for gray-scale images." Pattern Recognition Letters 17, no. 5 (1996): 437-450.
- [4] Criminisi, Antonio, Toby Sharp, and Andrew Blake. "Geos: Geodesic image segmentation." ECCV, 2008.
- [5] Wang, Guotai, et al. "
DeepIGeoS: A deep interactive geodesic framework for medical image segmentation." TPAMI, 2018.
A comparison of fast marching and ranster scan for 2D geodesic distance transform. (d) shows the Euclidean distance and (e) is a mixture of Geodesic and Euclidean distance.
This toolkit provides a cpp implementation of fast marching and raster scan for 2D/3D geodesic and
Euclidean distance transforms and a mixture of them, and proivdes a python interface to use it. This is part of the work of DeepIGeoS [5]. If you use our code, please cite this paper.
How to install
-
Install this toolkit easily by typing
pip install GeodisTK -
Alternatively, if you want to build from source files, download this package and run the following:
python setup.py build
python setup.py install
How to use
-
See a 2D example, run
python demo2d.py -
See a 3D example, run
python demo3d.py
Metadata
Release files for GeodisTK 0.1.7
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| GeodisTK-0.1.7.tar.gz | 9.6 kB | Without provenance | Details for GeodisTK-0.1.7.tar.gz |
Release files / GeodisTK-0.1.7.tar.gz
| Download URL | GeodisTK-0.1.7.tar.gz |
|---|---|
| Size | 9.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
282bd11f556a6044bf2ff2c4837e4cc01a1cdc0f990404997f83720dcba7c7d4
|
|
BLAKE2b-256 checksum How to use checksums |
0c4b214006a4675f6a2d5c78c0ad9cd78ec5d64c055f2a4f12db8c47d840abd1
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/46.1.3 requests-toolbelt/0.9.1 tqdm/4.40.2 CPython/3.6.9
|
Security for GeodisTK 0.1.7
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.