search

canismarko / xanespy

Chemical imaging of materials using X-ray absorbance spectroscopy
http://xanespy.readthedocs.io/en/latest/?badge=latest
GNU General Public License v3.0
4 stars 8 forks source link

Implement a mechanism for producing masks based on simple image contrast #6

Open WilliamJudge94 opened 5 years ago

WilliamJudge94 commented 5 years ago

Using the O K-edge with some ptychography data produces masks that are inverted (maybe?) from what they should be (i.e. 1's mark active material instead of background). We need to review the various edge combinations to make sure they're at least all consistent.

Also, there should be a page in the docs about using masks.

canismarko commented 5 years ago

Turns out this is the result of the underlying data, the edge jump happens to be relatively high in the background. This motivated the development of a second form of masking, named (contrast mask). @WilliamJudge94 is developing this.

canismarko commented 5 years ago

Code was completed with 0646af2f389b9636b7125da58a63c1f3b253a4ba, but the documentation needs a little cleaning up. https://github.com/canismarko/xanespy/commit/40f9779c804c0dba354d83ed56e072ac7e8f8236 added documentation on how to use the contrast mask, and lists one drawback of the absorption edge masking approach. We should list the benefits of XAS edge masking and the benefits and drawbacks of contrast masking.