sergiotarxz/libvips
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Comments for lbb resampler

Browse Source
This commit is contained in:
Nicolas Robidoux 2010-03-23 19:10:02 +00:00
parent 7f8938ae6c
commit 06f61bcf58
1 changed files with 64 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

72
libvips/resample/lbb.cpp
View File

@ -1,4 +1,4 @@
/* locally bounded bicubic resampler
/* lbb (locally bounded bicubic) resampler
*/
/*
@ -29,13 +29,70 @@
*/
/*
* 2009-2010 (c) Nicolas Robidoux, John Cupitt, Chantal Racette.
* 2010 (c) Nicolas Robidoux, John Cupitt, Chantal Racette.
*
* Nicolas Robidoux thanks Ralf Meyer, Minglun Gong, Adam Turcotte,
* Eric Daoust, Øyvind Kolås, Geert Jordaens, and Sven Neumann for
* useful comments and code.
*/
/*
* LBB (Locally Bounded Bicubic) is a high quality nonlinear variant
* of Catmull-Rom. Compared to Catmull-Rom, it produces resampled
* images with halos much reduced, both in terms of physical extent
* and over/undershoot amplitude. This is accomplished without
* noticeable changes to image smoothness.
*
* Another important property is that the resampled values are
* contained within the range of nearby input values.
*/
/*
* LBB is a novel method with the following properties:
*
* --When the limiters are inactive, it gives the same results as
* Catmull-Rom.
*
* --When used on binary images, in which case the limiters clamp
* everything to zero, LBB gives the same results as bicubic Hermite
* with all derivatives at the input pixel locations set to zero.
*
* --It is interpolatory.
*
* --It is C^1 with continuous cross derivatives.
*
* --It is locally bounded, in the following sense: Over each square
* patch, the surface is contained between the minimum and the
* maximum values among the 16 nearest input pixel values (those in
* the stencil).
*
* --It is globally bounded between the very smallest input pixel
* value and the very largest input pixel value. Consequently, it is
* not necessary to clamp results.
*
* --It is a Hermite bicubic method: The bicubic surface is defined,
* one convex hull of four nearby input points at a time, using the
* four point values, four x-derivatives, four y-derivatives, and four
* cross-derivatives.
*
* --The stencil for values in a square patch is the usual 4x4.
*
* --The LBB method is based on the method of Ken Brodlie, Petros
* Mashwama and Sohail Butt for constraining Hermite interpolants
* between globally defined planes:
*
* Visualization of surface data to preserve positivity and other
* simple constraints, Computer & Graphics, Vol. 19, #4, pages
* 585-594, 1995. DOI: 10.1016/0097-8493(95)00036-C.
*
* The main novelty of the LBB method (besides its reliance on slope
* limiters for image resampling) lies in the fact that the method
* of Brodlie et al is used to enforce local, as opposed to global,
* boundedness. This method was developed by Nicolas Robidoux and
* Chantal Racette of the Department of Mathematics and Computer
* Science of Laurentian University.
*/
#ifdef HAVE_CONFIG_H
#include <config.h>
#endif /*HAVE_CONFIG_H*/
@ -112,8 +169,8 @@ lbbicubic( const double c00,
/*
* STENCIL (FOOTPRINT) OF INPUT VALUES:
*
* The stencil of Symmetrized Monotone Catmull-Rom is the same as
* the standard Catmull-Rom's:
* The stencil of LBB is the same as for any standard Hermite
* bicubic (e.g., Catmull-Rom):
*
* (ix-1,iy-1) (ix,iy-1) (ix+1,iy-1) (ix+2,iy-1)
* = uno_one = uno_two = uno_thr = uno_fou
@ -444,11 +501,10 @@ lbbicubic( const double c00,
/*
* Call lbb with a type conversion operator as a parameter.
*
* It would be nice to do this with templates somehow---for one thing
* this would allow code comments!---but we can't figure a clean way
* to do it.
* It would be nice to do this with templates but we can't figure out
* how to do it cleanly. Suggestions welcome!
*/
#define LBB_CONVERSION( conversion ) \
#define LBB_CONVERSION( conversion ) \
template <typename T> static void inline \
lbb_ ## conversion( PEL* restrict pout, \
const PEL* restrict pin, \