I'm just getting into image processing and analysis using Python, and I'm looking for astronomical images (preferably raw, unprocessed images) to practice on. Can anyone suggest some places I can download public-access telescope images from?


I know you've already accepted an answer, but I feel that it doesn't provide a true representation of the type of image processing astronomers actually do. Actual image processing in astronomy is rarely done with JPG, PNG, or TIFF files. Instead, astronomers mostly work with FITS files and cameras/astronomical instruments actually produce FITS files when taking images. The advantage here is that FITS files can record much more information than just image information such as where/when the image was taken, multiple layers of the image, encoded WSC information, etc.

As part the mission objective of various organizations such as the National Science Foundation and NASA, all raw astronomical data taken in the United States by observatories funded by the United States are free and open to the public (after a limited time period in which the observing scientist gets exclusive rights to the data for their science objectives - usually several months to a year). This includes both ground (e.g. Keck, SDSS, Arecibo) and space observatories (e.g., Hubble or Spitzer). This is also true of many observatories funded by other countries as well.

A great place to start would be to head over to the Hubble Legacy Archive. Enter the site and try searching for an object or coordinates. For example, you can search for M101 (otherwise known as the Pinwheel Galaxy). You'll see a list of images come up and next to them, you can download the FITS file generated by Hubble when it took the image.

Another great resource would be the Sloan Digital Sky Survey. This is a systematic astronomical observation of as much of the sky as possible using a variety of instruments. You can practice looking at images and spectra. I should warn you that it will take a bit more effort to figure out where the data is and how to interpret it, but feel free to ask questions here when you run into trouble.

In general though, if you're looking for a particular type of image or data, find an observatory that records it and look for their publicly available data sets.

  • $\begingroup$ More directly for SDSS, you can search for object names or sky coordinates here, and get links to the (bzip2-compressed) FITS files: dr12.sdss.org/fields $\endgroup$ – Peter Erwin Aug 11 '17 at 14:03
  • 1
    $\begingroup$ Public availability of telescope data (after a proprietary period) isn't just a US thing: most major observatories worldwide do this, and have done so longer than the major US observatories have (except for Hubble). E.g., the European Southern Observatory, the Isaac Newton Observatory, the Anglo-Australian Telescope, etc. $\endgroup$ – Peter Erwin Aug 11 '17 at 14:05
  • 1
    $\begingroup$ A short, simple list of links to HST FITS files: spacetelescope.org/projects/fits_liberator/datasets_archives $\endgroup$ – Peter Erwin Aug 11 '17 at 14:06
  • 1
    $\begingroup$ @PeterErwin I know it's the case in the US for sure and knew it was true for other countries/observatories, but I didn't want to make a blanket statement without the proper research. $\endgroup$ – zephyr Aug 11 '17 at 14:07
  • 2
    $\begingroup$ To be really nitpicky, the availability of data from HST and other space observatories -- which are funded by NASA -- has nothing to do with the NSF. $\endgroup$ – Peter Erwin Aug 11 '17 at 14:23



Are two sites I have downloaded images from. I had to right click the image to download but it worked. Maybe more of a challenge for you are images of rain waves. I don't know where to find them though. I think all of images from above sites have been processed so they may not fit your needs.

These sites may help you.



The last site is for amateur image processors and has access to raw image data.

Good Luck.

  • 2
    $\begingroup$ Neither of the first two sites in that list provide links to "raw, unprocessed images", which would be in FITS format. $\endgroup$ – Peter Erwin Aug 11 '17 at 13:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.