Hi, I adapted the 2D code from: https://github.com/scikit-image/scikit-image/blob/master/skimage/feature/tem... to 3D: https://github.com/kevin-keraudren/template-matching-3d I use it in my work, I have manual segmentation experts performed on cropped data, and I need the template matching to realign the cropped image with the full image (3D medical images). So thanks a lot for the 2D code! If you think it is of use, I could fork scikit-image and add the 3D version next to the 2D version. I understand there should be as little duplicated code as necessary, but shall I create a function match_template3D or shall match_template check the shape of the input image? Kind regards, Kevin
Support for 3D images would be great! Imo checking for 3D images in the same function is the best option. Johannes Am 24.11.2013 um 23:45 schrieb Kevin Keraudren <kevin.keraudren@googlemail.com>:
Hi,
I adapted the 2D code from: https://github.com/scikit-image/scikit-image/blob/master/skimage/feature/tem...
to 3D: https://github.com/kevin-keraudren/template-matching-3d
I use it in my work, I have manual segmentation experts performed on cropped data, and I need the template matching to realign the cropped image with the full image (3D medical images). So thanks a lot for the 2D code!
If you think it is of use, I could fork scikit-image and add the 3D version next to the 2D version.
I understand there should be as little duplicated code as necessary, but shall I create a function match_template3D or shall match_template check the shape of the input image?
Kind regards,
Kevin
-- You received this message because you are subscribed to the Google Groups "scikit-image" group. To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image+unsubscribe@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
On Mon, Nov 25, 2013 at 10:41 AM, Johannes Schönberger <jsch@demuc.de>wrote:
Imo checking for 3D images in the same function is the best option.
+1 Sorry if I betray my ignorance about template matching, but can the 2D case not be considered a special case of the 3D with z=1? In other code (such as SLIC), when we get a 2D image we simply go: im = im[np.newaxis, ...] and feed that to the 3D code. Can the same be done for template matching? Thanks for this, Kevin! Juan.
participants (3)
-
Johannes Schönberger -
Juan Nunez-Iglesias -
Kevin Keraudren