Skip to content

cwru-sdle/pointextract

Repository files navigation

pointextract

Polar to cartesian transforms using annular point sampling

Designed to unwrap 2D cross section images of 3D X-ray computed tomography scans. The topological transformation enables the surface of a circular or elliptical object to be aligned for downsteam analysis.

Installation

You can install the package with:

pip install pointextract

Example

Simple example:

import pointextract
from skimage import io, filters

img_arr = io.imread('./data/sample.png')

img_thresh = img_arr > filters.threshold_otsu(img_arr)
ellipse = ellipse_detect(img_thresh)

img_unwrap = unwrap_image(img_arr, ellipse, radial_distance=20, num_points=800)

Questions

This package is still in early development. Please feel free to post to the GitHub Issues page with questions.

Acknowledgements

This material is based upon research in the Materials Data Science for Stockpile Stewardship Center of Excellence (MDS3-COE).

[Case Western Reserve University, SDLElab] 1