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

use 2d tables for bilinear

This commit is contained in:
John Cupitt 2008-10-28 21:04:52 +00:00
parent ae1763ae50
commit 1e120eb24e
3 changed files with 39 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
ChangeLog
View File

@ -4,6 +4,7 @@
- re-added type.[hc]
- added vipsinterpolate and im_affinei
- added yafr-smooth interpolation
- added yafrtest
11/9/08 started 7.16.3
- oop typo in manpage for im_project()

4
include/vips/interpolate.h
View File

@ -163,8 +163,8 @@ typedef struct _VipsInterpolateBilinearClass {
* to short), and double (for all others). We go to scale + 1, so
* we can round-to-nearest safely.
*/
int matrix_int[VIPS_TRANSFORM_SCALE + 1][2];
double matrix_double[VIPS_TRANSFORM_SCALE + 1][2];
int matrixi[VIPS_TRANSFORM_SCALE + 1][VIPS_TRANSFORM_SCALE + 1][4];
double matrixd[VIPS_TRANSFORM_SCALE + 1][VIPS_TRANSFORM_SCALE + 1][4];
} VipsInterpolateBilinearClass;
GType vips_interpolate_bilinear_get_type( void );

65
libsrc/mosaicing/interpolate.c
View File

@ -278,15 +278,10 @@ vips_interpolate_nearest_static( void )
#define BILINEAR_INT( TYPE ) { \
TYPE *tq = (TYPE *) q; \
\
const int m1 = class->matrix_int[xi][0]; \
const int m2 = class->matrix_int[xi][1]; \
const int m3 = class->matrix_int[yi][0]; \
const int m4 = class->matrix_int[yi][1]; \
\
const int c1 = (m3 * m1) >> VIPS_INTERPOLATE_SHIFT; \
const int c2 = (m3 * m2) >> VIPS_INTERPOLATE_SHIFT; \
const int c3 = (m4 * m1) >> VIPS_INTERPOLATE_SHIFT; \
const int c4 = (m4 * m2) >> VIPS_INTERPOLATE_SHIFT; \
const int c1 = class->matrixi[xi][yi][0]; \
const int c2 = class->matrixi[xi][yi][1]; \
const int c3 = class->matrixi[xi][yi][2]; \
const int c4 = class->matrixi[xi][yi][3]; \
\
const TYPE *tp1 = (TYPE *) p1; \
const TYPE *tp2 = (TYPE *) p2; \
@ -303,15 +298,10 @@ vips_interpolate_nearest_static( void )
#define BILINEAR_FLOAT( TYPE ) { \
TYPE *tq = (TYPE *) q; \
\
const double m1 = class->matrix_double[xi][0]; \
const double m2 = class->matrix_double[xi][1]; \
const double m3 = class->matrix_double[yi][0]; \
const double m4 = class->matrix_double[yi][1]; \
\
const double c1 = m3 * m1; \
const double c2 = m3 * m2; \
const double c3 = m4 * m1; \
const double c4 = m4 * m2; \
const double c1 = class->matrixd[xi][yi][0]; \
const double c2 = class->matrixd[xi][yi][1]; \
const double c3 = class->matrixd[xi][yi][2]; \
const double c4 = class->matrixd[xi][yi][3]; \
\
const TYPE *tp1 = (TYPE *) p1; \
const TYPE *tp2 = (TYPE *) p2; \
@ -392,7 +382,7 @@ vips_interpolate_bilinear_class_init( VipsInterpolateBilinearClass *class )
{
VipsInterpolateClass *interpolate_class =
(VipsInterpolateClass *) class;
int x;
int x, y;
vips_interpolate_bilinear_parent_class =
g_type_class_peek_parent( class );
@ -402,18 +392,35 @@ vips_interpolate_bilinear_class_init( VipsInterpolateBilinearClass *class )
/* Calculate the interpolation matricies.
*/
for( x = 0; x < VIPS_TRANSFORM_SCALE + 1; x++ ) {
/* At x == 0, we want to give all the weight to the LH pixel.
*/
const double c2 = (double) x / VIPS_TRANSFORM_SCALE;
const double c1 = 1.0 - c2;
for( x = 0; x < VIPS_TRANSFORM_SCALE + 1; x++ )
for( y = 0; y < VIPS_TRANSFORM_SCALE + 1; y++ ) {
double X, Y, Xd, Yd;
double c1, c2, c3, c4;
class->matrix_double[x][0] = c1;
class->matrix_double[x][1] = c2;
/* Interpolation errors.
*/
X = (double) x / VIPS_TRANSFORM_SCALE;
Y = (double) y / VIPS_TRANSFORM_SCALE;
Xd = 1.0 - X;
Yd = 1.0 - Y;
class->matrix_int[x][0] = c1 * VIPS_INTERPOLATE_SCALE;
class->matrix_int[x][1] = c2 * VIPS_INTERPOLATE_SCALE;
}
/* Weights.
*/
c1 = Xd * Yd;
c2 = X * Yd;
c3 = Xd * Y;
c4 = X * Y;
class->matrixd[x][y][0] = c1;
class->matrixd[x][y][1] = c2;
class->matrixd[x][y][2] = c3;
class->matrixd[x][y][3] = c4;
class->matrixi[x][y][0] = c1 * VIPS_INTERPOLATE_SCALE;
class->matrixi[x][y][1] = c2 * VIPS_INTERPOLATE_SCALE;
class->matrixi[x][y][2] = c3 * VIPS_INTERPOLATE_SCALE;
class->matrixi[x][y][3] = c4 * VIPS_INTERPOLATE_SCALE;
}
}
static void