Japanese grids publised by GSI could legally be included, pending conversion... about proj-data HOT 12 OPEN

I tried to convert Japanese par files to NTv2 files.

https://github.com/tohka/JapanGridShift

from proj-data.

@rouault

Sorry I've been so busy with my work that I haven't been able to make progress on this task.

You can add the gsb files converted from the parameter files published by the GSI to the repository or convert them to a new format.

• original parameter files
  • https://www.gsi.go.jp/sokuchikijun/tky2jgd_download.html
  • https://www.gsi.go.jp/sokuchikijun/sokuchikijun41012.html
• gsb files
  • https://github.com/tohka/JapanGridShift

They are available in accordance with the Geospatial Information Authority of Japan Website Terms of Use (compatible with CC BY 4.0).

• https://www.gsi.go.jp/kikakuchousei/kikakuchousei40182.html [in Japanese]
• https://www.gsi.go.jp/ENGLISH/page_e30286.html [in English]

It is important to note that we need to apply patches for earthquakes other than the Tohoku earthquake, and that in some areas, the changes are so complex that parameter files are not provided.

from proj-data.

I tried to convert Japanese par files to NTv2 files.

Cool ! Did you get a chance to check the result of the conversion of a few test points with what GSI perhaps publishes or other tools ?

from proj-data.

I checked some points.
For example, a point on 38.546789N 140.026224E in Tokyo Datum (Bessel) was transformed to 38.5497034780N 140.0228535294E in JGD2000 (GRS80) with Proj and tky2jgd.gsb.
It was transformed to 38.549703478N 140.022853531E with the web interface by GSI.

There is a slight difference between the results of the two methods and the cause is unclear.

from proj-data.

There is a slight difference

Unless I'm wrong, the difference is ~ 0.14 milimetre, whereas the precision of the transformation is reported to have a 20 cm accuracy by EPSG, so that's really neglectable. The different could result in floating-point inaccuracies in the computation chains of the GSI tool and PROJ, or using a different type of interpolation between grid values.

from proj-data.

@tohka Let us know when you feel that the grids can be included

from proj-data.

@tohka Hello. Do you think the grids are OK to be included ?

from proj-data.

Is there an open issue in PROJ-data related to this? I've been working in converting the GSIGEO2011 data from here into a proper tif. It seems I have reasonable results, but I haven't verified the precision yet, only compared it to EGM2008 for one transformation. Apart from that, I have some questions about the areas of use (I dropped data) and the central epoch of the grid file and I would like to ask them in the proper ticket.

from proj-data.

I have created a draft PR to add part of the Japanese geoid #37.

from proj-data.

@tohka, one colleague of mine contacted GSI to ask about which copyright applies to include GSIGEO2011 in a commercial application and she was asked to submit an application for requesting permission.
I'm the meantime I had made a draft PR including the .tif I made from gsigeo2011_ver2_1.asc, which we downloaded from https://fgd.gsi.go.jp/download/geoid.php after registering GSI's website. Now I'm wondering whether I should close it due to the copyright.
The Terms of Use are compatible with CC BY 4.0, but there is an exception for "duplication and use of Public Survey Results based on Survey Act". I'm a bit confused, is the geoid data from gsigeo2011_ver2_1.asc different from the data you used to create your .gsb files? It would be quite arbitrary that the geoid is protected but not the other files.

from proj-data.

@hernando
Under Japanese surveying law, both commercial and non-commercial use and reproduction of the results of public surveying must be applied for and approved.
https://www.gsi.go.jp/ENGLISH/page_e30029.html
In exceptional cases, no application is required for private use, for use by a limited number of people, or for use as a map where the precision of the illustration has been lost.
https://www.gsi.go.jp/common/000223838.pdf [in Japanese]

The Fundamental Geospatial Data that includes administrative boundaries, roads, rivers, geoids, etc. is available online, but since they are the results of public surveying, permission is required to use them for certain purposes. For example, if you want to use it on a publicly visible bulletin board, you must apply for a permit.

The files I converted are the data of variation due to the earthquake. When I asked the GSI, they replied that the data itself is not a public survey result and can be redistributed according to the rules (CC BY 4.0).

You can redistribute the geoid data by submitting an application and attaching a registration number.

from proj-data.

Thanks for the detailed reply. Then, as I understand it, the draft PR I submitted (#37), should be given up. I had already removed the file just in case, now I will close it.
I wonder if this website http://www.isgeoid.polimi.it/Geoid/Asia/Japan/japan11_g.html is sort of an exception, even if it's old data.

from proj-data.