# I The use of 3D reconstruction techniques to build a galaxy model

#### sbleuz

I'm learning about imaging techniques and projection/deprojection for images of galaxies at university. The big issue there is the fact that an image is a 2D representation of a 3D object (for example a galaxy), and to learn some properties of the object you need the 3D structure. What happens now is modeling, so basically you start from a model (for example spherical symmetry) and you try to calculate what properties the object has based on that simulation and you enhance the model as much as possible. What I am thinking of is completely replacing the model with an actual 3D image. Recently, even some smartphones have gotten the capability of making 3D pictures, using multiple camera's located next to each other on the back of the phone. Even though the camera's are quite close to each other, they're apparently far enough apart to take the picture from different angles, allowing for 3D modelling, which uses the slightly different points of origin to render a 3D image. I was thinking of applying this technique to say, a galaxy. Instead of using two camera's/ telescopes close together, i would use the same telescope at different times during the year. This way, the parallax of the earth would provide the different angles for the 3D model. What I want to know is if it would work. Would the parallax of the earth provide an angle big enough to do 3D modelling with? Or would a bigger angle be needed for that, say, the parallax of Mars or Jupiter maybe? Or maybe a smaller angle suffices, like the parallax due to the radius of the earth? I am looking forward to hear you opinion on this matter, or maybe even some arguments for or against the idea.

P.S. Apparently the title got messed up, it's probably too long or something. It should've said: "The use of 3D reconstruction techniques to build galaxy models".It's not that much of a difference, but if you thought it was weird, this is why.

Last edited:
Related Astronomy and Astrophysics News on Phys.org

#### websterling

First, Welcome to Physics Forums!

Sorry, but I don't think this is feasible. Just consider the distances involved. The Earth's orbit has a diameter of about 16 light minutes. Sirius, at a distance of ~8.6 light years, has a parallax of ~0.38 arcseconds. On galactic scales, the Andromeda galaxy is about 2.2 million light years distance, about 256000 times farther. My rough calculation gives it a parallax of ~1.6e-6 arcseconds. I don't think even the highest precision methods available could measure this. Even from Jupiter it wouldn't be enough to measure.

#### Drakkith

Staff Emeritus
2018 Award
What I want to know is if it would work. Would the parallax of the earth provide an angle big enough to do 3D modelling with? Or would a bigger angle be needed for that, say, the parallax of Mars or Jupiter maybe? Or maybe a smaller angle suffices, like the parallax due to the radius of the earth? I am looking forward to hear you opinion on this matter, or maybe even some arguments for or against the idea.
The Gaia spacecraft can supposedly measure the distance to stars with an accuracy of 10% out to the galactic core. You'd probably need to expand the spacecraft's orbit out to Mars or Jupiter to increase the accuracy significantly, but I'm exactly sure how the accuracy changes with the orbital radius.

https://en.wikipedia.org/wiki/Gaia_(spacecraft)

#### sbleuz

Ok, so apparently this is already being done inside our own galaxy, where distances are huge, but still ok for parallax, so it's possible to calculate positions from those. I did the math for the Andromeda galaxy and with the radius of the earths yearly cycle, we would need a telescope that can separate stars within 1/1000 to 1/10000 of an arcsecond. A little research tells me that right now, we can get to a few arcseconds, but that's it. So it will probably not be possible, unless someone invents a telescope without diffraction limit, which would break physics.

#### alantheastronomer

The Gaia spacecraft can supposedly measure the distance to stars with an accuracy of 10% out to the galactic core. You'd probably need to expand the spacecraft's orbit out to Mars or Jupiter to increase the accuracy significantly, but I'm exactly sure how the accuracy changes with the orbital radius.

https://en.wikipedia.org/wiki/Gaia_(spacecraft)
The distance to the galactic center is roughly 30 kilo-light-years, while the distance to the Andromeda galaxy is 2.5 mega-light-years or 100 times further, so you'd need a telescope at a radius 100 times further away, which is somewhere out in the Kuiper belt; Jupiter is only 5 times further away...

"The use of 3D reconstruction techniques to build a galaxy model"

### Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving