sis-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Martin Desruisseaux (JIRA)" <j...@apache.org>
Subject [jira] [Resolved] (SIS-176) Provide an optimized MathTransform implementation for diagonal matrices
Date Thu, 12 Nov 2015 14:07:10 GMT

     [ https://issues.apache.org/jira/browse/SIS-176?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Martin Desruisseaux resolved SIS-176.
-------------------------------------
    Resolution: Fixed

Implemented {{ScaleTransform}} for the diagonal matrix and for dropping trailing ordinate
values.

We do not provide a {{TranslateTransform}} implementation because rarely have only translations.
Even in the the _"Geocentric translations (geocentric domain)"_ case (EPSG:1031), we have
scale coefficients because of the change in ellipsoid axis lengths.

We do not provide an implementation for the scale + translation case because the performance
benefit compared to {{ProjectiveTransform}} is uncertain given the small size of typical matrices.



> Provide an optimized MathTransform implementation for diagonal matrices
> -----------------------------------------------------------------------
>
>                 Key: SIS-176
>                 URL: https://issues.apache.org/jira/browse/SIS-176
>             Project: Spatial Information Systems
>          Issue Type: Improvement
>          Components: Referencing
>    Affects Versions: 0.6
>            Reporter: Martin Desruisseaux
>            Assignee: Martin Desruisseaux
>            Priority: Minor
>             Fix For: 0.7
>
>
> {{ProjectiveTransform}} handles arbitrary matrices in a generic way. However in practice,
many matrices are diagonal + a translation transform. We may get some performance improvement
if we write a specialized transform class for such matrices, because it would replace a O(_n_²)
operation by a O(_n_) one (_n_ is the matrix size).



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message