[
https://issues.apache.org/jira/browse/SIS-599?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Martin Desruisseaux closed SIS-599.
-----------------------------------
> Robinson projection
> -------------------
>
> Key: SIS-599
> URL: https://issues.apache.org/jira/browse/SIS-599
> Project: Spatial Information Systems
> Issue Type: Sub-task
> Components: Referencing
> Reporter: Martin Desruisseaux
> Assignee: Martin Desruisseaux
> Priority: Major
> Fix For: 1.5
>
>
> Add support for the Robinson map projection. This projection is not described
> by EPSG guidance note, and I didn't not found it in Snyder's book neither.
> The only freely available reference that I found was
> [Wikipedia|https://en.wikipedia.org/wiki/Robinson_projection], but that
> reference gives no details about the interpolation method. The non-free
> resources used for this task is:
> Snyder, J. P. (1990). +The Robinson projection: A computation algorithm.+
> Cartography and Geographic Information Systems, 17 (4), p. 301-305.
> The Snyder's article gives a program in the ANSI Fortran language. That
> program has been translated to Java in this class and adapted for SIS
> architecture (constants moved to normalization matrices). Robinson did not
> specify a particular interpolation method, but Snyder's program uses the
> Stirling's central-difference formula, while PROJ uses cubic splices. Some
> other software use Aitken interpolation. The Fortran program translated to
> Java did not converged well for the inverse projection of points near a pole
> (maybe it is the reason why PROJ uses cubic splines instead). We modified the
> algorithm with the use of derivative (∂y/∂φ) for faster convergence.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)