Re: Making algorithms at least 3D, preferably nD
Volumetric image processing is definitely within scope of the scikit. The reason that most of the implementations up to this point were 2-D is simply a lack of time and hands. Luckily, that seems to be changing, so we may very well have the luxury of tackling this problem head-on.
I also think that the majority of use cases is based on 2-D data (plus channel data) and volumentric data is a specific and rare use case. I'm also dealing with lots of nD data (3-D, 4-D,…), nevertheless they are mostly still 2-D data. E.g. - SAR stacks (NxMxD) - Covariance matrix images (NxMxDxD) - Hyperspectral remote sensing images (NxMxD) - Tomographic SAR (NxMxJxD) - etc. where D is often > 200.
Volumetric 3D is rare to some, I'm sure ;) but there's a large audience available for medical image processing that needs 3D. Nearly everything I do involves 3D volumetric arrays. Confocal microscopy can generate true 3D multispectral data in the biological sciences. The field of neurology may be poised to have a huge need for true 3D (multispectral) processing in the near future as well! If you haven't seen this development out of Stanford, it's worth checking out: http://med.stanford.edu/ism/2013/april/clarity.html So, wherever possible, I'll be looking to add 3D implementations. I feel this is a "build it and they will come" sort of situation. On Monday, April 29, 2013 12:16:29 PM UTC-5, Johannes Schönberger wrote:
Volumetric image processing is definitely within scope of the scikit. The reason that most of the implementations up to this point were 2-D is simply a lack of time and hands. Luckily, that seems to be changing, so we may very well have the luxury of tackling this problem head-on.
I also think that the majority of use cases is based on 2-D data (plus channel data) and volumentric data is a specific and rare use case.
I'm also dealing with lots of nD data (3-D, 4-D,…), nevertheless they are mostly still 2-D data. E.g. - SAR stacks (NxMxD) - Covariance matrix images (NxMxDxD) - Hyperspectral remote sensing images (NxMxD) - Tomographic SAR (NxMxJxD) - etc. where D is often > 200.
Josh, thanks for the link. I'm not in this field but it looks quite interesting. Johannes Schönberger Am 30.04.2013 um 03:47 schrieb "Josh Warner" <silvertrumpet999@gmail.com<mailto:silvertrumpet999@gmail.com>>: Volumetric 3D is rare to some, I'm sure ;) but there's a large audience available for medical image processing that needs 3D. Nearly everything I do involves 3D volumetric arrays. Confocal microscopy can generate true 3D multispectral data in the biological sciences. The field of neurology may be poised to have a huge need for true 3D (multispectral) processing in the near future as well! If you haven't seen this development out of Stanford, it's worth checking out: http://med.stanford.edu/ism/2013/april/clarity.html So, wherever possible, I'll be looking to add 3D implementations. I feel this is a "build it and they will come" sort of situation. On Monday, April 29, 2013 12:16:29 PM UTC-5, Johannes Schönberger wrote:
Volumetric image processing is definitely within scope of the scikit. The reason that most of the implementations up to this point were 2-D is simply a lack of time and hands. Luckily, that seems to be changing, so we may very well have the luxury of tackling this problem head-on.
I also think that the majority of use cases is based on 2-D data (plus channel data) and volumentric data is a specific and rare use case. I'm also dealing with lots of nD data (3-D, 4-D,…), nevertheless they are mostly still 2-D data. E.g. - SAR stacks (NxMxD) - Covariance matrix images (NxMxDxD) - Hyperspectral remote sensing images (NxMxD) - Tomographic SAR (NxMxJxD) - etc. where D is often > 200. -- 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<mailto:scikit-image+unsubscribe@googlegroups.com>. For more options, visit https://groups.google.com/groups/opt_out.
participants (2)
-
Johannes Schönberger -
Josh Warner