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

use float div for sigma calc in gaussblur

Browse Source 
thanks aferrero2707
This commit is contained in:
John Cupitt 2014-04-21 19:24:14 +01:00
parent 35cea5dbf8
commit 03ad54361a
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
libvips/convolution/gaussblur.c
View File

@ -73,7 +73,7 @@ vips_gaussblur_build( VipsObject *object )
/* Stop at 20% of max ... bit mean, but means mask radius is roughly
* right.
*/
if( vips_gaussmat( &t[0], gaussblur->radius / 2, 0.2,
if( vips_gaussmat( &t[0], gaussblur->radius / 2.0, 0.2,
"separable", TRUE,
"integer", gaussblur->precision != VIPS_PRECISION_FLOAT,
NULL ) )
@ -162,8 +162,8 @@ vips_gaussblur_init( VipsGaussblur *gaussblur )
* This operator runs vips_gaussmat() and vips_convsep() for you on an image.
*
* @radius is not used directly. Instead the standard deviation of
* vips_gaussmat() is set to @radius / 2 and the minimum amplitude set to 20%.
* This gives a mask radius of approximately @radius pixels.
* vips_gaussmat() is set to @radius / 2.0 and the minimum amplitude set to
* 20%. This gives a mask radius of approximately @radius pixels.
*
* See also: vips_gaussmat(), vips_conv().
*