Make calculation conform to figure in https://arxiv.org/abs/2403.08538 #17
Conversation
Previously, the camera length was measured from the focus point, not the specimen plane as written in the paper
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #17 +/- ##
========================================
Coverage ? 100.00%
========================================
Files ? 5
Lines ? 190
Branches ? 21
========================================
Hits ? 190
Misses ? 0
Partials ? 0 ☔ View full report in Codecov by Sentry. |
| y_px, x_px : float | ||
| Detector pixel coordinates to project |
There was a problem hiding this comment.
As these are float, does (0.0, 0.0) correspond to the center of the top-left pixel, or to the top-left corner of the top-left pixel?
There was a problem hiding this comment.
Or is (0, 0) actually the "central pixel" of the detector, which is then actually the center of that pixel? That's what I gather from the "straight through" test case below...
| detector_pixel_size, scan_pixel_size : float | ||
| Physical pixel sizes. This assumes a uniform scan and detector grid in x | ||
| and y direction | ||
| camera_length : float | ||
| Virtual distance from specimen to detector | ||
| overfocus : float | ||
| Virtual distance from focus point to specimen. Underfocus is specified | ||
| as a negative overfocus. |
There was a problem hiding this comment.
What's the unit of these - probably all of these just have to be given in the same unit, but units don't matter otherwise?
| LiberTEM. It can be calculated with :fun:`get_transformation_matrix`. | ||
| fov_size_y, fov_size_x : float | ||
| Size of the scan area (field of view) in scan pixels. The scan | ||
| coordinate system is centered in the middle of the field of view. |
There was a problem hiding this comment.
What exactly does "centered" mean here? Where does the scan coordinate system have its origin?
| y_px, x_px : float | ||
| Detector pixel coordinates to project | ||
| cy, cx : float | ||
| Detector center in detector pixel coordinates |
There was a problem hiding this comment.
What does "center" mean here? (h/2, w/2)? Or the pixel which the optical axis is contained in?
| 'fov_size_y': 0, | ||
| 'fov_size_x': 0, |
There was a problem hiding this comment.
What does it mean to have fov_size_{x,y}=0? Shouldn't the scan have at least one scan pixel?
tests/test_overfocus.py
Outdated
| detector_shape=(detector_size, detector_size), | ||
| sim_params=params, | ||
| ) | ||
| # Center of the scan, severy second detector pixel |
There was a problem hiding this comment.
| # Center of the scan, severy second detector pixel | |
| # Center of the scan, every second detector pixel |
There was a problem hiding this comment.
Thanks for the updates - the doc string and comments make it a lot more clear! I've added some more questions - I think some of them can be answered by looking at the figure from the paper, on the other hand, with a bit more clear language it can stand on its own.
Thx! :-)
Previously, the camera length was measured from the focus point, not the specimen plane as written in the paper