I was playing around with the New NASA pic of earth (posted July 20th 2015 on Nasa Goddard space flight center on Flickr). I've found this very odd structure on the bottom right corner that is pretty much hidden in the darkness. There must be some kind of un-spectacular reason for this to look like a barber's pole, henceforth, this image will be named "Nasa's barber pole" until proven through some ultra-complicated computer mathematics.

enter image description here

I'm asking photographers because I would like a point of view from people who work with images either through Photoshop or whatever, Are these type of anomalies normal? Has anyone else noticed these things?

  • \$\begingroup\$ Do you have the link to the original? \$\endgroup\$
    – mattdm
    Jul 21, 2015 at 14:33
  • \$\begingroup\$ flic.kr/p/vj7kj2, it alsom mentions nothing about JPEG compression or any JPEG for that matter. this is the original pics "EXIF"-JFIFVersion - 1.01 X-Resolution - 72 dpi Y-Resolution - 72 dpi Orientation - Horizontal (normal) Color Space - sRGB \$\endgroup\$
    – NormLDude
    Jul 21, 2015 at 15:44
  • 1
    \$\begingroup\$ "JFIF" is a standard format for JPEG files. Details here. \$\endgroup\$
    – mattdm
    Jul 21, 2015 at 16:01
  • \$\begingroup\$ Ok, so JFIF is related to JPEG. Why do we get mediocre quality then, this is NASA we are talking about. And we are talking about an image of earth, last image of earth (The blue marble) is from 1972, so yup, I would like some real quality images. \$\endgroup\$
    – NormLDude
    Jul 21, 2015 at 16:08
  • 2
    \$\begingroup\$ I'm not sure I understand your expectations, here. Or why you're looking on Flickr, for that matter. Get the original 1977 blue marble image from visibleearth.nasa.gov/view.php?id=55418 - also JPEG, but these artifacts aren't visible. Or an updated composite image from 1997, available in TIFF formta: visibleearth.nasa.gov/view.php?id=54388 \$\endgroup\$
    – mattdm
    Jul 21, 2015 at 16:22

1 Answer 1


You're looking at JPEG artifacts. The JPEG compression scheme divides an image up into 8x8 pixel blocks and rebuilds each block using a collection of 2D waves as building blocks:

You can faithfully recreate any image by adding together a combination of wave images with the correct brightnesses. However JPEG is a lossy compression algorithm and so it throws away some waves to reduce size. Thus JPEG artifacts often resemble the more primitive waves as the higher order more complicated waves have been dropped. This can appear give structure to what might just be random noise in the original image.

Here is an extremely highly compressed image:

Most 8x8 blocks feature only the most primitive wave, the one in the top left corner of the first image. However in the background and around the "JPG" text you can see a few of the more detailed waves being used.

Your "barbers pole" is probably nothing more than a collection of dim stars in the background that has taken on a more structured, detailed looking appearance due to certain wave shapes being used that make it appear stripy with a ball on the top!

Either that, or it's aliens who can camouflage their ships to look exactly like JPEG artifacts in order to escape detection...

  • 4
    \$\begingroup\$ Why do you think the stars are removed? Dynamic range is dynamic range, even in space! \$\endgroup\$
    – mattdm
    Jul 21, 2015 at 15:59
  • 4
    \$\begingroup\$ @NormLDude they posted it on Flickr which obviously isn't their official repository for their data. NASA does have publicly accessible websites where you can download the raw data in the form of text files containing the measured gray values of the pixels, see e.g. here. \$\endgroup\$ Jul 21, 2015 at 16:20
  • 3
    \$\begingroup\$ @NormLDude Why do you keep saying "they removed the stars"? \$\endgroup\$
    – mattdm
    Jul 21, 2015 at 16:23
  • 2
    \$\begingroup\$ They're just a combination of noise in the image and jpeg artefacts. They might be interesting to you, but unless you're the type of person who hears hooves and expects to see zebra or really loves the fine artistry of the discrete cosine transform there is nothing of interest there. \$\endgroup\$ Jul 21, 2015 at 16:35
  • 2
    \$\begingroup\$ @NormLDude a simple paragraph would have sufficed under the pic that was posted on Flickr "May not be exactly as shown here" or "The quality of this image is 44% of the original" - the artifacts you're referring to are completely invisible until you perform a huge shadow boost, so why on Earth would they need to mention it? Should the same disclaimer be placed next to every JPEG image published anywhere in the web? Images NASA posts on Flickr are for PR purposes, not scientific analysis. \$\endgroup\$
    – Matt Grum
    Jul 22, 2015 at 9:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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