Allow array alpha for imshow #8183
Conversation
|
Thanks for submitting this, it looks like a good improvement! A couple of things that need to be fixed:
There's lots of information that should cover the above at http://matplotlib.org/devel/index.html - let us know if you need any help with anything though. |
| @@ -419,14 +428,18 @@ def _make_image(self, A, in_bbox, out_bbox, clip_bbox, magnification=1.0, | |||
| # colormapping works correctly | |||
| hid_output = output | |||
| output = np.ma.masked_array( | |||
| hid_output[..., 0], hid_output[..., 3] < 0.5) | |||
| hid_output[..., 0], hid_output[..., 3] == 0) | |||
There was a problem hiding this comment.
Yes, this line is indeed causing the problem. In particular, the alpha channel is interpolated near the boundary of an image. The < 0.5 implementation will set pixels that are "more than half transparent" to fully transparent whereas the == 0 implementation only treats them as masked if they are fully transparent.
In short, the boundaries in a few tests differ slightly.
I believe the == 0 implementation is preferable because it effectively provides anti aliasing on the boundary by restoring the alpha channel. Here is a zoomed-in example.
There was a problem hiding this comment.
Hm, but this causes trouble if there are bad data in image. For example, bad values in one of the tests are shown as blue. If we restore the alpha channel, all bad blue values will become transparent. As a compromise, I have rerendered the problematic images (changes only occur at the boundary) and I only restore the alpha channel if _array_alpha is given.
|
I have to run to dinner, but I suspect the |
|
Please rebase instead of merging, and don't make extra empty commits just to trigger CI. |
|
FYI, we (members of the matplotlib organisation) can manually trigger CI if needed, and you should be able to manually trigger it on your own fork. |
|
Ok, will do that in the future. |
|
Please leave this one for me to review / merge. |
|
This looks useful, but fell under the radar. @tillahoffmann do you have the bandwidth to rebase, and we can try and resurrect? If no one touches it for a while, please ping (and re-ping). Sorry for the lack of response here. |
|
Will close this one and send a new PR because the implementation has changed enough to make a new implementation easier. |
|
Don’t be shy at harassing us |
|
🐑 I thought this had gone in already! This one is definitely on me... |
This PR allows the
alphaargument ofimshowto be an array with shape matching the image to support pixel-by-pixel opacity.