Solved

Reprojection is off

  • 13 June 2018
  • 8 replies
  • 11 views

Badge

I am quite new to FME, so sorry for my lack of knowledge. I've created a very basic workspace to combine three LAS point clouds, reproject them from State Plane Alaska Zone 4 Feet to UTM84-6N. The combination and reprojection work and create a file, but the new LAS file is now georeferenced about 300 kms east of where it should be. I've done a similar transformation in LASTools successfully. Any ideas?

icon

Best answer by daveatsafe 18 June 2018, 23:19

View original

8 replies

Userlevel 2
Badge +17

Hi @dvandommelen,

That distance is about right for a NAD27/NAD83 shift. FME has coordinate system definitions for both NAD27 and NAD83 in Zone 4.

If you have manually set the input coordinate system, please check the datum on it. If FME is reading the coordinate system from the LAS files, try overriding it by setting the input coordinate system.

If neither of these work, please let me know, and I will escalate this into a support case.

Badge

Hi @dvandommelen,

That distance is about right for a NAD27/NAD83 shift. FME has coordinate system definitions for both NAD27 and NAD83 in Zone 4.

If you have manually set the input coordinate system, please check the datum on it. If FME is reading the coordinate system from the LAS files, try overriding it by setting the input coordinate system.

If neither of these work, please let me know, and I will escalate this into a support case.

Thanks. I thought that I’d set it for NAD83 but I might have missed that. I will try to reproject on Thursday.

 

 

Badge

Hi @dvandommelen,

That distance is about right for a NAD27/NAD83 shift. FME has coordinate system definitions for both NAD27 and NAD83 in Zone 4.

If you have manually set the input coordinate system, please check the datum on it. If FME is reading the coordinate system from the LAS files, try overriding it by setting the input coordinate system.

If neither of these work, please let me know, and I will escalate this into a support case.

@DaveAtSafe I've been working on this a bit today. I've very carefully checked and I have been using NAD83, which is what is also specified by the survey company that produced the LiDAR. I'm fairly certain that my specification of the coordinate system is correct.

 

 

In checking the various FME coordinate systems, I did find that there are actually two different versions of the Alaska State Plane Zone 4 NAD83 listed. One is called AK83-4 (AK83-4F) and the other is listed by EPSG: 26934. The latter seems to be available only by meters. I note that they have different eastings listed in their parameters, which seems significant to me.

 

 

I attempted to create a custom coordinate system using the EPSG: 26934, but it kept returning it with meters, though I had specified feet. I'll try editing it in a text reader, next, but I'm no whiz at these sorts of things.

 

Badge

Hi @dvandommelen,

That distance is about right for a NAD27/NAD83 shift. FME has coordinate system definitions for both NAD27 and NAD83 in Zone 4.

If you have manually set the input coordinate system, please check the datum on it. If FME is reading the coordinate system from the LAS files, try overriding it by setting the input coordinate system.

If neither of these work, please let me know, and I will escalate this into a support case.

I double-checked and the eastings are consistent, my mistake. I also checked the OGC descriptions and everything seems to check out.

 

 

Userlevel 2
Badge +17
I double-checked and the eastings are consistent, my mistake. I also checked the OGC descriptions and everything seems to check out.

 

 

Hi @dvandommelen,

 

Would you mind sharing one of the LAS files with me? If you don't want to make it public, you can send a download link (ie. Google Drive, Dropbox) to dave.campanas@safe.com, and I will try to find a reason for the offset.
Badge

@DaveAtSafe Will do on Monday, I’m headed out for a long weekend. Thanks for your help.

Userlevel 2
Badge +17

Hi @dvandommelen,

I looked at your .LAS file, and the coordinate system read by FME seems to be incorrect, with a meters based X offset for a foot based coordinate system.

If you override the input coordinate system, either in the Navigator pane of Workbench, or in the settings for the CsmapReprojector transformer, with the coordinate system AK83-4F, the point cloud will be properly reprojected to UTM84-6N.

In the CsmapReprojector, please also set the Vertical Handling to 'Heights are relative to the ellipsoid(s) or geocentric'. This will ensure the elevations are converted to meters along with the XY values.

Badge

Hi @dvandommelen,

I looked at your .LAS file, and the coordinate system read by FME seems to be incorrect, with a meters based X offset for a foot based coordinate system.

If you override the input coordinate system, either in the Navigator pane of Workbench, or in the settings for the CsmapReprojector transformer, with the coordinate system AK83-4F, the point cloud will be properly reprojected to UTM84-6N.

In the CsmapReprojector, please also set the Vertical Handling to 'Heights are relative to the ellipsoid(s) or geocentric'. This will ensure the elevations are converted to meters along with the XY values.

That worked perfectly!

 

 

Reply