core/vil/algo/vil_dog_pyramid.h File Reference

Compute a pyramid of difference of gaussian images. More...

#include <vil/algo/vil_gauss_filter.h>
#include <vil/vil_resample_bilin.h>
#include <vil/vil_math.h>
#include <vcl_vector.h>

Go to the source code of this file.

Functions
template<class T >
void	vil_dog_pyramid (const vil_image_view< T > &src_image, vcl_vector< vil_image_view< T > > &smooth_pyramid, vcl_vector< vil_image_view< T > > &dog_pyramid, unsigned min_size)
	Compute a pyramid of difference of gaussian images.

---

Detailed Description

Compute a pyramid of difference of gaussian images.

Author:

Tim Cootes

Definition in file vil_dog_pyramid.h.

---

Function Documentation

template<class T >

void vil_dog_pyramid	(	const vil_image_view< T > &	src_image,
		vcl_vector< vil_image_view< T > > &	smooth_pyramid,
		vcl_vector< vil_image_view< T > > &	dog_pyramid,
		unsigned	min_size
	)

Compute a pyramid of difference of gaussian images.

Computes each layer of a pyramid by smoothing then computing the difference from the original image. The smoothed is then subsampled using a reduction factor of 1.5 (ie each level is 2/3 the size of the level below) and used to produced the next level.

min_size defines the smallest dimension (restricting the number of levels that will be constructed)

This is useful for finding locally interesting points and their associated scales - see "Object Recognition from Scale Invariant Features" D.Lowe, ICCV1999, pp.1150-1157.

Definition at line 29 of file vil_dog_pyramid.h.