1. Mick West

    Mick West Administrator Staff Member

    Nobody would be more excited than me if someone found evidence of life on Mars. But I really don't think this is it:

    71 49'19.73"N 29 33'06.53"W

    [​IMG]

    Supposedly it's something that looks like a base on Mars. It's superficially like a long white structure, with various buildings.

    The problem here is one of pixels. What the above image is is a highly magnified and filtered version something that's just about 13 pixels long. It's like taking a tiny spec on some photo, and blowing it up as much as possible, and then smoothing out all the edges, blurring, and smoothing again.

    Here' it more zoomed out

    [​IMG]

    And here with the contrast enhanced

    [​IMG]

    If you look around at the surrounding pixels, it seems clear that any of the "structure" of the object probably just comes from the noise.

    You'll note the background has a faint grid visible. This is from the compression, and the squares here represent 8x8 pixel squares in the original image.

    To me this looks like a glitch in the data where a few pixels were set to random bright white and dark colors, and that has been compounded by the image being rotated (to fit the mapping onto the globe), filtered, and compressed, to give this result. Here's an example of what might happen:

    We start off with a high quality picture straight from the camera. This is photo of the desert to the East of the Grand Canyon, on Earth.

    [​IMG]

    I've reddend and darkened it a bit, so this is like the original photo from Mars. On the right is a zoomed in section of this, showing the individual pixels.

    [​IMG]

    Let's introduce a data transmission glitch that results in a few pixels in word (16 pixels) being set to white pixels. You'll see it as a thin horizontal line one the right of the image.

    [​IMG]

    Now to simulate the transform and compression process, I'll rotate it eight degrees, then save it out with high compression. Jpeg Level 3 in photoshop:

    [​IMG]

    Then we map it onto a bit of Mars in Google Earth, with "Add Image Overlay"
    [​IMG]

    Then we zoom in using Google Earth.
    [​IMG]

    And here's another example with different simulated glitches. One line is solid white, the other is random white and transparent:

    [​IMG]

    So yeah, maybe it's it space station on Mars. But a tiny little bit of corrupt data seems a lot more likely.
    Last edited: Nov 20, 2013
  2. Mick West

    Mick West Administrator Staff Member

    Last edited: Nov 20, 2013
  3. Mick West

    Mick West Administrator Staff Member

    Here's one scientist's theory on what caused the original glitch

    http://www.foxnews.com/scitech/2011...ase-found-on-google-mars-debunked/?test=faces

    Basically it's a cosmic ray, an energetic charged subatomic paticel , that zappet though the camera, and caused momentary excitation of the CCD sensor, causing a few pixels to be brightened. It's like video static, except there's only one little bit of it, because it was caused by one particle.

    I'm not entirely convinced. It looks more like corrupt data that a streak of light, especially as it's pixel aligned. I suppose it could be related to how the CDD is scanned.

    [​IMG]
    Last edited: Nov 20, 2013
  4. Mick West

    Mick West Administrator Staff Member

  5. Mick West

    Mick West Administrator Staff Member

    Last edited: Nov 20, 2013

Share This Page