lti::ogd2Kernel< T > Class Template Reference

Second order Oriented Gaussian Derivative. More...

Inheritance diagram for lti::ogd2Kernel< T >:

Collaboration diagram for lti::ogd2Kernel< T >:

Public Member Functions
	ogd2Kernel ()
	ogd2Kernel (const int &size, const double &theVariance=1.4426950409, const double &theAngle=0.0)
void	generate (const int &size, const double &theVariance, const double &theAngle)

Detailed Description

Second order Oriented Gaussian Derivative.

OGD kernels are separable! (one filter pair per basis kernel)

The first basis filter of the second order OGD is:

$b_0(x,y) = g(x,y) \frac{x^2 - \sigma^2}{\sigma^4}$

and the second:

$b_1(x,y) = g(x,y) \frac{y^2 - \sigma^2}{\sigma^4}$

and the third:

$b_2(x,y) = g(x,y) \frac{xy}{\sigma^4}$

The used interpolation functions are $\cos^2\theta$ , $\sin^2\theta$ and $2\cos\theta \sin\theta$ .

template<class T >

lti::ogd2Kernel< T >::ogd2Kernel ( )

Default constructor.

Creates an empty kernel. To initialize use generate() or generateBasisKernel().

template<class T >

Constructor.

Parameters:

	size	is the dimension of the one dimensional part (i.e. the kernel has dimensions size x size !)
	theVariance	variance of the kernel
	theAngle	used angle (in radians)

template<class T >

void lti::ogd2Kernel< T >::generate	(	const int &	size,
		const double &	theVariance,
		const double &	theAngle
	)

Initialize this kernel with the specified values.

Parameters:

	size	is the dimension of the one dimensional part (i.e. the kernel has dimensions size x size !)
	theVariance	variance of the kernel
	theAngle	used angle (in radians)

The documentation for this class was generated from the following file: