Peter Williams
GitHub
Коммит
a5b13ff19d
|
|
@ -1,4 +1,4 @@
|
|||
# 0.1.0dev0 (unreleased)
|
||||
# 0.1.0.dev0 (unreleased)
|
||||
|
||||
- No changes yet
|
||||
|
||||
|
|
|
|||
172
README.md
172
README.md
|
|
@ -6,12 +6,11 @@
|
|||
|
||||
<!--pypi-begin-->
|
||||
[toasty] is a Python library that helps you create [TOAST] “tile pyramids”
|
||||
from [FITS] astronomical image files. These multi-resolution all-sky maps can
|
||||
be viewed in software such as the [AAS] [WorldWide Telescope].
|
||||
from astronomical image data. These multi-resolution all-sky maps can be
|
||||
viewed in software such as the [AAS] [WorldWide Telescope].
|
||||
|
||||
[toasty]: https://toasty.readthedocs.io/
|
||||
[TOAST]: https://doi.org/10.3847/1538-4365/aaf79e
|
||||
[FITS]: https://en.wikipedia.org/wiki/FITS
|
||||
[AAS]: https://aas.org/
|
||||
[WorldWide Telescope]: http://www.worldwidetelescope.org/
|
||||
|
||||
|
|
@ -30,7 +29,37 @@ The easiest way to install [toasty] is through [pip]:
|
|||
pip install toasty
|
||||
```
|
||||
|
||||
[pip]: https://pypi.org/project/pip/
|
||||
[pip]: https://pip.pypa.io/
|
||||
|
||||
For more information, please see the full [toasty installation instructions].
|
||||
|
||||
[toasty installation instructions]: https://toasty.readthedocs.io/en/latest/installation.html
|
||||
|
||||
|
||||
## Documentation, Examples, etc.
|
||||
|
||||
For documentation and examples, go to:
|
||||
|
||||
https://toasty.readthedocs.io/
|
||||
|
||||
|
||||
## Contributions
|
||||
|
||||
Contributions to [toasty] are welcome! See
|
||||
[the WorldWide Telescope contributors’ guide] for applicable information. We
|
||||
use a standard workflow with issues and pull requests. All participants in
|
||||
[toasty] and the WWT communities must abide by the [WWT Code of Conduct].
|
||||
|
||||
[the WorldWide Telescope contributors’ guide]: https://worldwidetelescope.github.io/contributing/
|
||||
[WWT Code of Conduct]: https://worldwidetelescope.github.io/code-of-conduct/
|
||||
|
||||
|
||||
## Release History
|
||||
|
||||
Releases of [toasty] are logged in the file [CHANGES.md](./CHANGES.md), as
|
||||
well as release listings maintained by
|
||||
[GitHub](https://github.com/WorldWideTelescope/toasty/releases) and
|
||||
[PyPI](https://pypi.org/project/toasty/#history).
|
||||
|
||||
|
||||
## Dependencies
|
||||
|
|
@ -53,137 +82,12 @@ pip install toasty
|
|||
[pytest]: https://docs.pytest.org/
|
||||
|
||||
|
||||
### Usage
|
||||
## Legalities
|
||||
|
||||
```python
|
||||
from toasty import toast
|
||||
|
||||
toast(data_sampler, depth, base_dir, wtml_file=None, merge=True, base_level_only=False,
|
||||
ra_range=None, dec_range=None, toast_tile=None, restart=False, top_layer=0)
|
||||
```
|
||||
|
||||
where:
|
||||
|
||||
* **data_sampler** function or string
|
||||
- A function of (lon, lat) that samples a datasetat the input 2D coordinate arrays
|
||||
- A string giving a base toast directory that contains the base level of toasted tiles, using this option, only the merge step takes place, the given directory must contain a "depth" directory for the given depth parameter
|
||||
* **depth** int
|
||||
- The depth of the tile pyramid to create (4^d tiles are created at a depth of d)
|
||||
* **base_dir** str
|
||||
- The path to create the files at
|
||||
* **wtml_file** str (optional)
|
||||
- The path to write a WTML file to. If not present, no file will be written
|
||||
* **merge** bool or callable (default True)
|
||||
How to treat lower resolution tiles.
|
||||
- If True, tiles above the lowest level (highest resolution) will be computed by averaging and downsampling the 4 subtiles.
|
||||
- If False, sampler will be called explicitly for all tiles
|
||||
- If a callable object, this object will be passed the 4x oversampled image to downsample
|
||||
* **base_level_only** bool (default False)
|
||||
If True only the bottem level of tiles will be created.
|
||||
In this case merge will be set to True, but no merging will happen, and only the highest resolution layer of images will be created.
|
||||
* **ra_range** array (optional)
|
||||
* **dec_range** array (optional)
|
||||
To toast only a portion of the sky give min and max ras and decs ([minRA,maxRA],[minDec,maxDec]) in degrees.
|
||||
If these keywords are used base_level_only will be automatically set to true, regardless of its given value.
|
||||
* **toast_tile** array\[n,x,y\] (optional)
|
||||
If this keyword is used the output will be all the subtiles of toast_tile at the given depth (base_level_only will be automatically set to true, regardless of its given value.
|
||||
* **top_layer** int (optional)
|
||||
If merging this indicates the uppermost layer to be created.
|
||||
[toasty] is copyright Chris Beaumont, Clara Brasseur, and the AAS WorldWide
|
||||
Telescope Team. It is licensed under the MIT License.
|
||||
|
||||
|
||||
## Acknowledgments
|
||||
|
||||
|
||||
Toasty provides a few basic sampler functions:
|
||||
|
||||
* **healpix_sampler** for sampling from healpix arrays
|
||||
* **cartesian_sampler** for sampling from cartesian-projections
|
||||
* **normalizer** for applying an intensity normalization after sampling
|
||||
|
||||
### Examples
|
||||
|
||||
To toast an all-sky, Cartesian projection, 8 byte image:
|
||||
|
||||
```python
|
||||
from toasty import toast, cartesian_sampler
|
||||
from skimage.io import imread
|
||||
|
||||
data = imread('allsky.png')
|
||||
sampler = cartesian_sampler(data)
|
||||
toastDirectory = 'toast'
|
||||
depth = 8 # approximately 0.165"/pixel at highest resolution
|
||||
|
||||
# Getting the full toast tile set for the entire sky
|
||||
toast(sampler, depth, toastDirectory)
|
||||
|
||||
# Toasting a specific region of the sky defined by ra/dec bounds
|
||||
raRange = [208.8,212.2]
|
||||
decRange = [52.5,56.8]
|
||||
toast(sampler, depth, outtoastDirectoryput, ra_range=raRange, dec_range=decRange)
|
||||
|
||||
# Toasting a specific region of the sky defined by a higher level toast tile
|
||||
tile=[4,5,9]
|
||||
toast(sampler, depth, toastDirectory, toast_tile=tile)
|
||||
|
||||
# Creating only the bottom layer of toast tiles
|
||||
toast(sampler, depth, toastDirectory,base_level_only=True)
|
||||
|
||||
# Merging from a previously created toast layer up to a specified top layer
|
||||
topLayer = 4
|
||||
toast(toastDirectory, depth, toastDirectory, top_layer=topLayer)
|
||||
```
|
||||
|
||||
To apply a log-stretch to an all sky FITS image:
|
||||
|
||||
```python
|
||||
from toasty import toast, cartesian_sampler, normalizer
|
||||
from astropy.io import fits
|
||||
|
||||
data = fits.open('allsky.fits')[0].data
|
||||
vmin, vmax = 100, 65535
|
||||
scaling = 'log'
|
||||
contrast = 1
|
||||
sampler = normalizer(cartesian_sampler(data), vmin, vmax
|
||||
scaling, bias, contrast)
|
||||
output = 'toast'
|
||||
depth = 8
|
||||
toast(sampler, depth, output)
|
||||
```
|
||||
|
||||
To perform a custom transformation
|
||||
|
||||
```python
|
||||
from toasty import toast
|
||||
from astropy.io import fits
|
||||
|
||||
data = fits.open('allsky.fits')[0].data
|
||||
|
||||
def sampler(x, y):
|
||||
"""
|
||||
x and y are arrays, giving the RA/Dec centers
|
||||
(in radians) for each pixel to extract
|
||||
"""
|
||||
... code to extract a tile from `data` here ...
|
||||
|
||||
output = 'toast'
|
||||
depth = 8
|
||||
toast(sampler, depth, output)
|
||||
```
|
||||
|
||||
See ``toasty.tile`` for documentation on these functions.
|
||||
|
||||
### Using with WorldWide Telescope
|
||||
To quickly preview a toast directory named `test`, navigate to the directory
|
||||
where `test` exists and run
|
||||
|
||||
```
|
||||
python -m toasty.viewer test
|
||||
```
|
||||
|
||||
This will start a web server, probably at [http://0.0.0.0:8000](http://0.0.0:8000) (check the output for the actual address). Open this URL in a browser to get a quick look at the data.
|
||||
|
||||
For more information about using WorldWide Telescope with custom image data,
|
||||
see [the official documentation](http://www.worldwidetelescope.org/Docs/worldwidetelescopedatafilesreference.html). The function `toasty.gen_wtml` can generate the wtml information for images generated with toasty.
|
||||
|
||||
For an example of tiles generated with Toasty, see [The ADS All Sky Survey](http://adsass.org/wwt). The code used to generate these images is [here](https://github.com/ChrisBeaumont/adsass/blob/master/toast/toast.py).
|
||||
|
||||
For an example of tiles generated using the Toasty functionality specific to this fork see [The STScI AstroView Panstarrs Survey](https://mast.stsci.edu/portal/Mashup/Clients/AstroView/AstroView.html?debug&avSurveyType=PANSTARRS). The code used to generate these images is [here](https://github.com/ceb8/toastPanstarrs).
|
||||
Work on [toasty] is funded through the [AAS] [WorldWide Telescope] project.
|
||||
|
|
|
|||
|
|
@ -9,15 +9,16 @@ These are notes for the `toasty` developers about how to create a new release.
|
|||
5. For the final commit, update the version number in `setup.py` and
|
||||
`docs/conf.py`, and add a proper version and date to `CHANGES.md`. Commit
|
||||
with message `Release version ${version}`.
|
||||
6. Create a pull request on GitHub for the new release.
|
||||
6. Push to GitHub and create a pull request for the new release called
|
||||
"Release PR for version $version".
|
||||
7. Get it so that it passes CI, creating fixup commits as necessary.
|
||||
8. When it's really really ready, `git clean -fxd && python setup.py sdist
|
||||
bdist_wheel --universal && twine upload dist/*.tar.gz dist/*.whl`. If
|
||||
`twine` finds problems, make any final changes and retry.
|
||||
8. When it's really really ready, `git clean -fxd && python setup.py sdist &&
|
||||
twine upload dist/*.tar.gz`. If `twine` finds problems, make any final
|
||||
changes and retry.
|
||||
9. If needed, do a `git rebase -i` to make the version-bump commit the last
|
||||
one again.
|
||||
10. `git tag v${version}`
|
||||
11. Update the version number to `${cur_major}.${next_minor}.0dev0` and add a
|
||||
11. Update the version number to `${cur_major}.${next_minor}.0.dev0` and add a
|
||||
new separator in `CHANGES.md` along the lines of `${version} (unreleased)`.
|
||||
Commit with a message of `Back to development.`
|
||||
12. `git push` (with `-f` if history was rewritten) to the PR branch. This had
|
||||
|
|
|
|||
|
|
@ -4,8 +4,8 @@ project = 'toasty'
|
|||
author = 'Chris Beaumont and the AAS WorldWide Telescope Team'
|
||||
copyright = '2014-2019, ' + author
|
||||
|
||||
release = '0.1.0dev0' # the full version string; also update ../setup.py
|
||||
version = '0.1.0dev0' # the "short" version
|
||||
release = '0.1.0.dev0' # the full version string; also update ../setup.py
|
||||
version = '0.1.0.dev0' # the "short" version
|
||||
|
||||
extensions = [
|
||||
'sphinx.ext.autodoc',
|
||||
|
|
|
|||
|
|
@ -0,0 +1,137 @@
|
|||
=============================
|
||||
Examples of How to Use Toasty
|
||||
=============================
|
||||
|
||||
Here we’ll summarize some of the ways that you can use toasty_.
|
||||
|
||||
.. _toasty: https://toasty.readthedocs.io/
|
||||
|
||||
|
||||
Toasting a Cartesian, all-sky PNG image
|
||||
=======================================
|
||||
|
||||
This script processes an existing all-sky PNG image that uses a Cartesian
|
||||
projection, using scikit-image_ to load the data::
|
||||
|
||||
from toasty import toast, cartesian_sampler
|
||||
from skimage.io import imread
|
||||
|
||||
data = imread('allsky.png')
|
||||
sampler = cartesian_sampler(data)
|
||||
output_directory = 'toast'
|
||||
depth = 8 # approximately 0.165"/pixel at highest resolution
|
||||
|
||||
# Make the full toast tile set for the entire sky:
|
||||
toast(sampler, depth, output_directory)
|
||||
|
||||
.. _scikit-image: https://scikit-image.org/
|
||||
|
||||
|
||||
Toasting subsets of the sky
|
||||
===========================
|
||||
|
||||
You don’t have to build the full pyramid for the full sky all at once::
|
||||
|
||||
# Toast a specific region of the sky defined by RA/Dec bounds
|
||||
toast(sampler, depth, output_directory,
|
||||
ra_range = [208.8, 212.2], # degrees
|
||||
dec_range = [52.5, 56.8], # degrees
|
||||
)
|
||||
|
||||
# Toast a specific region of the sky defined by a higher-level TOAST tile
|
||||
toast(sampler, depth, output_directory,
|
||||
toast_tile = [4, 5, 9],
|
||||
)
|
||||
|
||||
# Create only the bottom layer of toast tiles
|
||||
toast(sampler, depth, output_directory,
|
||||
base_level_only = True,
|
||||
)
|
||||
|
||||
# Merge from a previously created toast layer up to a specified layer
|
||||
toast(sampler, depth, output_directory,
|
||||
top_layer = 4,
|
||||
)
|
||||
|
||||
|
||||
Controlling how data are turned into RGB
|
||||
========================================
|
||||
|
||||
Here we apply a log-stretch to an all sky FITS image::
|
||||
|
||||
from toasty import toast, cartesian_sampler, normalizer
|
||||
from astropy.io import fits
|
||||
|
||||
data = fits.open('allsky.fits')[0].data
|
||||
|
||||
vmin, vmax = 100, 65535
|
||||
scaling = 'log'
|
||||
contrast = 1
|
||||
sampler = normalizer(
|
||||
cartesian_sampler(data),
|
||||
vmin, vmax
|
||||
scaling, bias, contrast
|
||||
)
|
||||
|
||||
output_directory = 'toast'
|
||||
depth = 8
|
||||
toast(sampler, depth, output_directory)
|
||||
|
||||
|
||||
Non-Cartesian coordinate transformations
|
||||
========================================
|
||||
|
||||
A custom “sampler” can be used to tell toasty_ what image values
|
||||
correspond to what locations on the sky::
|
||||
|
||||
from toasty import toast
|
||||
|
||||
def sampler(x, y):
|
||||
"""
|
||||
x and y are arrays, giving the RA/Dec centers
|
||||
(in radians) for each pixel to extract
|
||||
"""
|
||||
... code to produce pixel values goes here ...
|
||||
|
||||
output_directory = 'toast'
|
||||
depth = 8
|
||||
toast(sampler, depth, output_directory)
|
||||
|
||||
See :meth:`toasty.tile` for documentation on sampler functions.
|
||||
|
||||
|
||||
Previewing toasts with AAS WorldWide Telescope
|
||||
==============================================
|
||||
|
||||
To quickly preview a toast directory named ``mytoast``, run the command::
|
||||
|
||||
python -m toasty.viewer mytoast
|
||||
|
||||
This will start a web server, probably at ``http://0.0.0.0:8000``. (Check the
|
||||
program’s output for the actual address). Open this URL in a browser to get a
|
||||
quick look at the data.
|
||||
|
||||
|
||||
Example toasty outputs: ADS All-Sky Survey
|
||||
==========================================
|
||||
|
||||
For an example of tiles generated with Toasty, see `the ADS All Sky Survey
|
||||
<http://adsass.org/wwt>`_. The code used to generate these images is in `the
|
||||
file toast.py`_ in the `adsass/wwt-frontend`_ repository on GitHub. This tile
|
||||
pyramid was created by Chris Beaumont.
|
||||
|
||||
.. _the file toast.py: https://github.com/adsass/wwt-frontend/blob/master/toast/toast.py
|
||||
.. _adsass/wwt-frontend: https://github.com/adsass/wwt-frontend/
|
||||
|
||||
|
||||
Example toasty outputs: PanSTARRS in STScI AstroView
|
||||
====================================================
|
||||
|
||||
Another example is a toasting of the “3π” survey of the PanSTARRS_ project,
|
||||
viewable `here in STScI’s AstroView`_ interface. The code used to generate
|
||||
these images is in the GitHub repository `ceb8/toastPanstarrs`_. This tile
|
||||
pyramid was created by Clara Brasseur.
|
||||
|
||||
.. _PanSTARRS: https://panstarrs.stsci.edu/
|
||||
.. _here in STScI’s AstroView: https://mast.stsci.edu/portal/Mashup/Clients/AstroView/AstroView.html?debug&avSurveyType=PANSTARRS
|
||||
.. _ceb8/toastPanstarrs: https://github.com/ceb8/toastPanstarrs
|
||||
|
|
@ -2,10 +2,15 @@
|
|||
toasty: Creating TOAST tile pyramids in Python
|
||||
==============================================
|
||||
|
||||
The ``toasty`` package helps you create `TOAST
|
||||
<https://arxiv.org/abs/1812.03926>`_ tile pyramids that can be used in
|
||||
software such as the `AAS <https://aas.org>`_ `WorldWide Telescope
|
||||
<http://worldwidetelescope.org/home>`_.
|
||||
toasty_ is a Python library that helps you create TOAST_ “tile pyramids”
|
||||
from astronomical image data. These multi-resolution all-sky maps can
|
||||
be viewed in software such as the AAS_ `WorldWide Telescope`_.
|
||||
|
||||
.. _toasty: https://toasty.readthedocs.io/
|
||||
.. _TOAST: https://arxiv.org/abs/1812.03926
|
||||
.. _AAS: https://aas.org/
|
||||
.. _WorldWide Telescope: http://www.worldwidetelescope.org/home
|
||||
|
||||
|
||||
User guide
|
||||
==========
|
||||
|
|
@ -13,19 +18,22 @@ User guide
|
|||
.. toctree::
|
||||
:maxdepth: 1
|
||||
|
||||
overview
|
||||
installation
|
||||
examples
|
||||
api
|
||||
|
||||
|
||||
Getting help
|
||||
============
|
||||
|
||||
If you run into any issues when using ``toasty``, please open an issue `on its
|
||||
If you run into any issues when using toasty_, please open an issue `on its
|
||||
GitHub repository <https://github.com/WorldWideTelescope/toasty/issues>`_.
|
||||
|
||||
|
||||
Acknowledgments
|
||||
===============
|
||||
|
||||
Work on ``toasty`` is funded through the American Astronomical Society
|
||||
(`AAS`_) `WorldWide Telescope`_ project. The first version of ``toasty`` was
|
||||
Work on toasty_ is funded through the American Astronomical Society
|
||||
(`AAS`_) `WorldWide Telescope`_ project. The first version of toasty_ was
|
||||
written by Chris Beaumont.
|
||||
|
|
|
|||
|
|
@ -0,0 +1,44 @@
|
|||
=================
|
||||
Installing toasty
|
||||
=================
|
||||
|
||||
Installing toasty with pip
|
||||
==========================
|
||||
|
||||
You can also install the latest release of toasty using pip_::
|
||||
|
||||
pip install toasty
|
||||
|
||||
.. _pip: https://pip.pypa.io/en/stable/
|
||||
|
||||
|
||||
Dependencies
|
||||
============
|
||||
|
||||
If you install toasty using pip_ as described above, any required dependencies
|
||||
will get installed automatically. The `README in the Git repository`_ lists the
|
||||
current dependencies if you would like to see an explict list.
|
||||
|
||||
.. _README in the Git repository: https://github.com/WorldWideTelescope/toasty/#readme
|
||||
|
||||
|
||||
Installing the developer version
|
||||
================================
|
||||
|
||||
If you want to use the very latest developer version, you should clone `this
|
||||
repository <https://github.com/WorldWideTelescope/toasty/>`_ and manually
|
||||
install the package in “editable” mode::
|
||||
|
||||
git clone https://github.com/WorldWideTelescope/toasty.git
|
||||
cd toasty
|
||||
pip install -e .
|
||||
|
||||
You will need to have Numpy_ and Cython_ installed on order to build the extension
|
||||
module that toasty uses to speed up its core operations.
|
||||
|
||||
.. _Numpy: https://cython.org/
|
||||
.. _Cython: https://numpy.org/
|
||||
|
||||
You can run the test suite with the command::
|
||||
|
||||
pytest toasty
|
||||
|
|
@ -0,0 +1,101 @@
|
|||
===========================
|
||||
Overview: What is Toasting?
|
||||
===========================
|
||||
|
||||
So, toasty_ is for toasting. What does that even mean?
|
||||
|
||||
.. _toasty: https://toasty.readthedocs.io/
|
||||
|
||||
This documentation will *not* go into detail because there are plans to
|
||||
provide comprehensive information in a separate document that is not specific
|
||||
to this particular library.
|
||||
|
||||
Alas, that documentation has not been written yet. In the meantime, the best
|
||||
detailed reference is `McGlynn et al. 2019`_, which explains and justifies the
|
||||
Tesselated Octahedral Adaptive Spherical Transformation (TOAST) projection
|
||||
that toasty_ is concerned with. See also the `WorldWide Telescope Projection
|
||||
Reference`_, although that document is a bit out of date. (Which shouldn’t
|
||||
matter, in principle, but its current Web expression may be a bit busted.) The
|
||||
following text aims to give a quick overview of the concepts that these
|
||||
documents work out in detail.
|
||||
|
||||
.. _McGlynn et al. 2019: https://ui.adsabs.harvard.edu/abs/2019ApJS..240...22M/abstract
|
||||
.. _WorldWide Telescope Projection Reference: https://worldwidetelescope.gitbook.io/projection-reference/
|
||||
|
||||
|
||||
The problem(s)
|
||||
==============
|
||||
|
||||
Say that you are writing a piece of software that aim to let users
|
||||
interactively explore a map of the Earth, the sky, or another planet. For the
|
||||
purposes of this discussion, the key thing is that these are all spherical
|
||||
entities. The user may want to zoom out and view large swathes of these
|
||||
spheres at once, or they may want to zoom way in and see extremely fine
|
||||
detail. The user will probably be accessing the maps over a network, and the
|
||||
maps may be far too large to transmit in over the network in their entirety.
|
||||
|
||||
There are two problems to solve here. First, we need some way to store image
|
||||
data covering an entire sphere. There is no unique best way to map the curved
|
||||
surface of a sphere into a two-dimensional representation; in particular, many
|
||||
common projections perform poorly at the poles.
|
||||
|
||||
Second, once we have some image data that we can map onto a sphere in some
|
||||
satisfactory way, we need a way to transmit pieces of it to the user
|
||||
incrementally. If they zoom out, it won't be practical to send them a huge
|
||||
chunk of the full-resolution map.
|
||||
|
||||
|
||||
The solution(s)
|
||||
===============
|
||||
|
||||
Our solution to the above problems is to construct “tile pyramids” using the
|
||||
TOAST spherical projection.
|
||||
|
||||
The TOAST projection does exactly what we need for the first problem: it maps
|
||||
the entire sphere onto a 2D image that can easily be represented digitally.
|
||||
While TOAST is not perfect in every way, it performs well at the poles and
|
||||
maintains approximately uniform resolution at all locations on the sphere. If
|
||||
you’re familiar with this topic, you may know that the HEALPix_ projection
|
||||
also operates in this problem space. We won’t go into details here, but
|
||||
suffice it to say that TOAST has some nice technical properties that make it
|
||||
the preferred choice for software such as the AAS_ `WorldWide Telescope`_. In
|
||||
particular, TOAST maps the sphere onto a square image, rather than the jagged
|
||||
shape required by HEALPix.
|
||||
|
||||
.. _HEALPix: https://healpix.jpl.nasa.gov/
|
||||
.. _AAS: https://aas.org/
|
||||
.. _WorldWide Telescope: http://www.worldwidetelescope.org/home
|
||||
|
||||
A high-resolution full-sphere TOAST map may weigh in at a million pixels *on a
|
||||
side*, or a trillion pixels total — far too large to manipulate in a
|
||||
user-facing scenario. Creating a “tile pyramid” of the image makes it possible
|
||||
to actually do useful things with a full-scale TOAST map. First, the
|
||||
high-resolution map is broken into a set of square sub-images, “tiles,” each
|
||||
256 pixels on a side. By the requirements of the TOAST standard, the number of
|
||||
tiles on each side must be a power of 2. Then, lower-resolution maps are
|
||||
created by downsampling the high-resolution map in 2×2 pixel blocks, yielding
|
||||
derived images that are half as large as their parents along each axis. (Each
|
||||
downsampled image therefore has 1/4 as many pixels as its parent.) By
|
||||
construction, each of these lower-resolution maps can also be broken into
|
||||
256×256 tiles. This process continues until a “level 0” map is created
|
||||
consisting of a single tile that represents the full sphere. With a full
|
||||
sphere having a solid angle of 4π steradians, each pixel in this level-0 map
|
||||
covers about 0.0002 steradians or 0.6 square degrees. As a user zooms in or
|
||||
out, or pans around the sphere, their computer is sent the tiles needed to
|
||||
produce a visually acceptable view. Each individual tile file is small enough
|
||||
that the data transfer remains tractable.
|
||||
|
||||
|
||||
The role of toasty
|
||||
==================
|
||||
|
||||
The toasty_ module helps create these TOAST tile pyramids from astronomical
|
||||
image data. There are essentially three problems that it tackles:
|
||||
|
||||
1. Transforming image data from their native projection to the TOAST one.
|
||||
2. Mapping scalar data values into RGB color images for user display.
|
||||
3. Downsampling the high-resolution TOAST map all the way down to the level-0
|
||||
map.
|
||||
|
||||
The process of creating TOAST tile pyramids from some input data is
|
||||
colloquially referred to as “toasting.”
|
||||
4
setup.py
4
setup.py
|
|
@ -34,8 +34,8 @@ project homepage].
|
|||
|
||||
setup_args = dict(
|
||||
name = 'toasty',
|
||||
version = '0.1.0dev0', # also update docs/conf.py
|
||||
description = 'Generate TOAST image tile pyramids from FITS files',
|
||||
version = '0.1.0.dev0', # also update docs/conf.py
|
||||
description = 'Generate TOAST image tile pyramids from existing image data',
|
||||
long_description = get_long_desc(),
|
||||
long_description_content_type = 'text/markdown',
|
||||
url = 'https://toasty.readthedocs.io/',
|
||||
|
|
|
|||
Загрузка…
Ссылка в новой задаче