# Images in physical space in Python

**URL:** <https://discourse.itk.org/t/images-in-physical-space-in-python/2124>\
**Category:** Engineering\
**Tags:** python\
**Created:** [August 5, 2019, 6:16pm UTC](https://discourse.itk.org/t/images-in-physical-space-in-python/2124 "2019-08-05T18:16:33Z")\
**Posts on this page:** 1\
**Showing post:** 23

<div class="post-metadata">

**Author:** ![blowekamp](https://discourse.itk.org/user_avatar/discourse.itk.org/blowekamp/32/79_2.png) [@blowekamp](https://discourse.itk.org/u/blowekamp)\
**Post date:** [August 15, 2019, 1:50am UTC](https://discourse.itk.org/t/images-in-physical-space-in-python/2124/23 "2019-08-15T01:50:53Z")

</div>

@matt.mccormick That is looking like a good start.

There was a [recent discussion](https://discourse.itk.org/t/making-a-3d-sequence-from-4d-volume/2070) of how to handle time. Similarly there are a couple different ways to order color. i.e. CXYZ, XYZTC etc. Simularly tensor images too. How can non-physical dimensions be handled? ITK uses a templated pixel type to separate a physical dimension form “pixel components”

There is also the definition and order of an “index”. We mean `(i,j,k)` where i is the fasted changing index, but then `ndarrays` are indexed via `[k,j,i]`.

I expect others from different domains will have additional concerns and comments to improve the specification too. It may be worthwhile to go through a more formal proposal and comment process to aid in getting more eyes on the spec, and interest.

Would the `LPS` direction imply that 2-D it just `LP` with the origin being in the top right?

---

_[View the full topic](https://discourse.itk.org/t/images-in-physical-space-in-python/2124)._