You are currently browsing the tag archive for the ‘Fourier Transform Secret Messages space Messier’ tag.

Messier Object 68

I was reading through my Digital Image Processing book (2nd edition) about something called Fourier transforms. Without explaining the math, suffice it to say that any 2d image can be transformed into the Fourier domain. When you do a Fourier transform on a normal image, you get what looks like a cluster of stars. So me being me, I thought, what do you get if you do a Fourier transform on a picture of stars? SECRET MESSAGES FROM SPACE!?

How to decode messages from space.
1. Find a good candidate photo from space. Should have a bright spot in the very center and be 512×512 pixels.
2. Convert the photo to floating point precision.
3. Perform an inverse Fourier transform (I used a shake plugin from pixelmaina), using the same image as both the real and mathematical “imaginary” portion (you know, the square root of -1) of the Fourier transform.
4. Center, Scroll, or otherwise (Filter>Other>Offset in Photoshop) the resulting image so that the brightest part is in the center.
5. Interpret the results.

Messier Object 68 in Fourier space

Messier Object 68 in Fourier space after recentering

Results:*

Messier Object 5 in Fourier Space

Messier object 5 (M5) clearly shows that somewhere in the universe, a noisy television exists.

Messier Object 30 in Fourier Space

M30 likewise.

Messier Object 68 in Fourier space after recentering

M68 shows a planet with 50 moons and a race of sentient dogs.

*Final interpretations will be subjected to peer review as of Aug 2090.

Areas for Further Research:
The three Messier objects I have thus far decoded clearly show a low signal-to-noise ratio. In analyzing other Fourier images, I note that bright pixel clusters are extremely rare. Rather, the image is comprised of individual pixels of varying brightnesses(eses). Thus, my photos of stars are not true Fourier images. For better results, I must use a diagram of the stars instead of a photo, with each star being a maximum of one pixel in diameter.