6 replies to this topic

[JerryYeh]

When NASA took the picture of the earth, they create a 3D model first , then use some algorithm to tranform the 3D into 2D map, I am working on something similiar but I do not have the software to do the transforming process, I wonder if there's any software or plug-in out there that can do the job for me? I've been doing some search on google but I found nothing so far, anyone can help me? Thanks.

[Hans van der Maarel]

When NASA took the picture of the earth, they create a 3D model first , then use some algorithm to tranform the 3D into 2D map, I am working on something similiar but I do not have the software to do the transforming process, I wonder if there's any software or plug-in out there that can do the job for me? I've been doing some search on google but I found nothing so far, anyone can help me? Thanks.

Not necessarily with a 3d model really, you can achieve it by simply georeferencing too. I use Global Mapper for this, but other software can do the trick too. You'll need a reference dataset and then identify control points (i.e. point A in your photo corresponds to point X in your reference data), repeat that for enough control points and Global Mapper will accurately* reproject the photo to the coordinate system of your reference dataset.

* accuracy depends on how accurate you can identify the control points and how oblique the photo is.

[JerryYeh]

When NASA took the picture of the earth, they create a 3D model first , then use some algorithm to tranform the 3D into 2D map, I am working on something similiar but I do not have the software to do the transforming process, I wonder if there's any software or plug-in out there that can do the job for me? I've been doing some search on google but I found nothing so far, anyone can help me? Thanks.

Not necessarily with a 3d model really, you can achieve it by simply georeferencing too. I use Global Mapper for this, but other software can do the trick too. You'll need a reference dataset and then identify control points (i.e. point A in your photo corresponds to point X in your reference data), repeat that for enough control points and Global Mapper will accurately* reproject the photo to the coordinate system of your reference dataset.

* accuracy depends on how accurate you can identify the control points and how oblique the photo is.

Thank you for your reply, but what i'm dealing with, is a sphere earth product , which a artist has done some artwork on it(he painted directly on a sphere), but we need the 2D version of the image, we 3D scanned the image into the computer but don't know how to flatten the image, I know this proberbly a little more about 3D than cartography, but
i wonder if you guys ever came across something that can do the job for me? thanks.

[Hans van der Maarel]

Thank you for your reply, but what i'm dealing with, is a sphere earth product , which a artist has done some artwork on it(he painted directly on a sphere), but we need the 2D version of the image, we 3D scanned the image into the computer but don't know how to flatten the image, I know this proberbly a little more about 3D than cartography, but
i wonder if you guys ever came across something that can do the job for me? thanks.

Okay, in what format do you have the 3D data? Can you get a single photo/render from one hemisphere?

[Michael Schmeling]

Thank you for your reply, but what i'm dealing with, is a sphere earth product , which a artist has done some artwork on it(he painted directly on a sphere), but we need the 2D version of the image, we 3D scanned the image into the computer but don't know how to flatten the image, I know this proberbly a little more about 3D than cartography, but
i wonder if you guys ever came across something that can do the job for me? thanks.

I assume you got the scanned data as a set of (x,y,z) coordinates. You can convert these coordinates into spherical coordinates (Wikipedia) and then use the theta and phi values to plot the image in 2D. I.e. a point with (theta,phi) coordinates on the sphere becomes a point with (x,y) coordinates in the image, where (x,y) is something like (theta/pi*width, phi/2pi*height).

The easiest was to do this is probably to write a little script.

[JerryYeh]

Thank you for your reply, but what i'm dealing with, is a sphere earth product , which a artist has done some artwork on it(he painted directly on a sphere), but we need the 2D version of the image, we 3D scanned the image into the computer but don't know how to flatten the image, I know this proberbly a little more about 3D than cartography, but
i wonder if you guys ever came across something that can do the job for me? thanks.

I assume you got the scanned data as a set of (x,y,z) coordinates. You can convert these coordinates into spherical coordinates (Wikipedia) and then use the theta and phi values to plot the image in 2D. I.e. a point with (theta,phi) coordinates on the sphere becomes a point with (x,y) coordinates in the image, where (x,y) is something like (theta/pi*width, phi/2pi*height).

The easiest was to do this is probably to write a little script.

Thank you for your reply, it helps alot, I guess i might have to do more searching or find some way to write a script myself.

[Michael Schmeling]

I can help you with it. Contact me privately if you want.