Image Component Library (ICL): icl::filter::WeightedSumOp Class Reference

Image Component Library (ICL)

All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines

Accumulate weighted pixel values of all image channels. More...

#include <WeightedSumOp.h>

Inheritance diagram for icl::filter::WeightedSumOp:

List of all members.

Public Member Functions
	WeightedSumOp ()
	creates a new WeightedSumOp object
	WeightedSumOp (const std::vector< icl64f > &weights)
	creates an new WeightedSumOp object with a given weights vector
void	apply (const core::ImgBase poSrc, core::ImgBase *ppoDst)
	applies this operation on the source image
const std::vector< icl64f > &	getWeights () const
	returns the current weight vector
void	setWeights (const std::vector< icl64f > &weights)
	sets up the current weights vector
Private Attributes
std::vector< icl64f >	m_vecWeights
	internal storage for the channel weights

Detailed Description

Accumulate weighted pixel values of all image channels.

Pixels of all channels in source image are weighted by a channel-wise weight and accumulated to the destination image:

$D(x,y,0) = \sum\limits_{c=0}^C S(x,y,c)*w(c)$

where $D$ is the destination image, $S$ is the source image, $w$ is the weights vector and $C$ is the source images channel count. The result image is created with depth32f by default. Only if the source image has a 64Bit floating point depth (depth64f), the destination image is adapted to depth64f to avoid loss of precession.

Performance: 1000x1000 image with 10 channels (1400Mhz Centrino)

icl8u: 82ms
icl32f: 96ms
icl64f: 170ms

Performance: 1000x1000 image with 3 channels (1400Mhz Centrino)

icl8u: 21ms
icl32f: 25ms
icl64f: 47ms

Constructor & Destructor Documentation

icl::filter::WeightedSumOp::WeightedSumOp ( ) [inline]

creates a new WeightedSumOp object

icl::filter::WeightedSumOp::WeightedSumOp ( const std::vector< icl64f > & weights ) [inline]

creates an new WeightedSumOp object with a given weights vector

Parameters:

weights channel weights vector

Member Function Documentation

void icl::filter::WeightedSumOp::apply	(	const core::ImgBase *	poSrc,
		core::ImgBase **	ppoDst
	)		`[virtual]`

applies this operation on the source image

Parameters:

poSrc	source image
ppoDst	destination image (adapted to icl32f by default, if the source image has depth64f, ppoDst is adapted to icl64f too.

Implements icl::filter::UnaryOp.

const std::vector<icl64f>& icl::filter::WeightedSumOp::getWeights ( ) const [inline]

returns the current weight vector

Returns:: reference to the current weight vector

void icl::filter::WeightedSumOp::setWeights ( const std::vector< icl64f > & weights ) [inline]

sets up the current weights vector

Parameters:

weights new weight vector

Member Data Documentation

std::vector<icl64f> icl::filter::WeightedSumOp::m_vecWeights [private]

internal storage for the channel weights

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

/home/jenkins/workspace/icl-manual-trunk/icl-manual/ICLFilter/src/ICLFilter/WeightedSumOp.h

All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines