Question

CSAR raster functionality to read/write measurement coordinate information (true position)

  • 17 October 2022
  • 3 replies
  • 1 view

Hello,

For a FME workflow involving hydrographic CSAR raster files both band and point coordinate information are present in the file. The point coordinate information is the location where the measurement to generate the raster file was taken. This is crucial information for a user.

According to the help page of CSAR raster (CARIS Spatial Archive (CSAR) Raster Quick Facts (safe.com)) this GCP and Z-value information, which to my knowledge is the point measurement location, is not included in the functionality of the CSAR reader or writer. Is there a workaround available in FME to include this point information in the CSAR raster file ? It is important that is saved in the same manner as in the original file. If not,would it be possible to add this functionality in future releases ?

 

Thank you!


3 replies

Userlevel 2
Badge +13

Hello @yasminvb​, FME supports georeferenced rasters, so this information should be discoverable. Unfortunately, we do not currently support GCP in the CSAR Raster Format, and I believe this may be a limitation of the library we're using under the hood (eg. does not appear to support GCP's).

 

Are you interested in georeferencing your dataset using GCP's or do you just want a georeferenced raster? If the latter, it should be possible. If I've misunderstood the ask, please feel free to re-iterate! Happy to help, Kailin.

Hello Kailin,

 

The client I am helping needs the point location where (hydrographic) measurements were done. They call it the "true position". So their CSAR raster export from CARIS contains both depth bands/layers and in addition also the "true position" of the depth band (deepest measurements) which is connected to the depth band. FME reads the "true position" as bands instead of coordinate (x,y,z) data. The CSAR raster writer does also write the "true position" coordinates as bands/layers instead of x,y,z coordinates. This is not what the user wants because it is not the way they expect the data to be structured.

 

I have added an attachment. In red is the original format and in blue is the way the CSAR raster file is structured after it passes through FME. You can see two bands with the coordinate information (true position) where added: Depth_pos.0 and Depth_pos.1. This is not what they want. They want to be able to click on the depth layer and then they want to get the coordinates in the green box. It seems that the CSAR reader and writer of FME are not able to maintain this information in the original data structure.

 

I hope my explanation is clear. :)

 

Kind regards,

Yasmin

 

 

image

Userlevel 2
Badge +13

Hello Kailin,

 

The client I am helping needs the point location where (hydrographic) measurements were done. They call it the "true position". So their CSAR raster export from CARIS contains both depth bands/layers and in addition also the "true position" of the depth band (deepest measurements) which is connected to the depth band. FME reads the "true position" as bands instead of coordinate (x,y,z) data. The CSAR raster writer does also write the "true position" coordinates as bands/layers instead of x,y,z coordinates. This is not what the user wants because it is not the way they expect the data to be structured.

 

I have added an attachment. In red is the original format and in blue is the way the CSAR raster file is structured after it passes through FME. You can see two bands with the coordinate information (true position) where added: Depth_pos.0 and Depth_pos.1. This is not what they want. They want to be able to click on the depth layer and then they want to get the coordinates in the green box. It seems that the CSAR reader and writer of FME are not able to maintain this information in the original data structure.

 

I hope my explanation is clear. :)

 

Kind regards,

Yasmin

 

 

image

Hello @yasminvb​, thank you for reiterating! This is a lot clearer, the image was helpful. Sorry you're encountering this. Would you be able to submit a support case with us, so we can take a closer look at your (input/output) datasets? You can use the online form: here. Happy to help, Kailin

Reply