lti::quadTreeSegmentation Class Reference

Quad-Tree based color image segmentation method. More...

#include <ltiQuadTreeSegmentation.h>

Inheritance diagram for lti::quadTreeSegmentation:

[legend]

Collaboration diagram for lti::quadTreeSegmentation:

[legend]

List of all members.

Classes
class	parameters
	The parameters for the class quadTreeSegmentation. More...
Public Member Functions
	quadTreeSegmentation ()
	quadTreeSegmentation (const parameters &par)
	quadTreeSegmentation (const quadTreeSegmentation &other)
virtual	~quadTreeSegmentation ()
virtual const char *	getTypeName () const
bool	apply (const image &src, image &dest) const
bool	apply (const image &src, imatrix &mask, palette &pal) const
bool	split (const image &img, imatrix &mask) const
quadTreeSegmentation &	copy (const quadTreeSegmentation &other)
quadTreeSegmentation &	operator= (const quadTreeSegmentation &other)
virtual functor *	clone () const
const parameters &	getParameters () const
Protected Member Functions
bool	stats (const image &img, const point &from, const point &to, trgbPixel< float > &stats) const
bool	split (const image &img, const float &threshold, const point &minSize, tree< rectangle >::node &qtree) const

---

Detailed Description

Quad-Tree based color image segmentation method.

This is a very simple but inefficient method to segment color images. It is provided because it is one of the classical methods of split-and-merge segmentation techniques.

A quad-tree based partition of the image if first done, and after that neighbor regions are merged if they are similar enough.

---

Constructor & Destructor Documentation

lti::quadTreeSegmentation::quadTreeSegmentation

(

)

Default constructor.

lti::quadTreeSegmentation::quadTreeSegmentation

(

const parameters &

par

)

Construct a functor using the given parameters.

lti::quadTreeSegmentation::quadTreeSegmentation

(

const quadTreeSegmentation &

other

)

Copy constructor.

Parameters:

other

the object to be copied

virtual lti::quadTreeSegmentation::~quadTreeSegmentation

(

)

[virtual]

Destructor.

---

Member Function Documentation

bool lti::quadTreeSegmentation::apply	(	const image &	src,
		imatrix &	mask,
		palette &	pal
	)			const

Generate a second color image, where each found region gets the mean color obtain from the original image.

This is not the real segmentation, but for visualization purposes some times this is the desired functionality.

Parameters:

	src	image with the source data.
	mask	labeled mask with the segmented regions.
	pal	color palette for each region (the mean value)

Returns:

true if apply successful or false otherwise.

bool lti::quadTreeSegmentation::apply	(	const image &	src,
		image &	dest
	)			const

Generate a second color image, where each found region gets the mean color obtain from the original image.

This is not the real segmentation, but for visualization purposes some times this is the desired functionality.

Parameters:

	src	image with the source data.
	dest	image where the result will be left.

Returns:

true if apply successful or false otherwise.

virtual functor* lti::quadTreeSegmentation::clone

(

)

const [virtual]

Returns a pointer to a clone of this functor.

Reimplemented from lti::segmentation.

quadTreeSegmentation& lti::quadTreeSegmentation::copy

(

const quadTreeSegmentation &

other

)

Copy data of "other" functor.

Parameters:

other

the functor to be copied

Returns:

a reference to this functor object

Reimplemented from lti::segmentation.

const parameters& lti::quadTreeSegmentation::getParameters

(

)

const

Returns used parameters.

Reimplemented from lti::functor.

virtual const char* lti::quadTreeSegmentation::getTypeName

(

)

const [virtual]

Returns the name of this type ("quadTreeSegmentation").

Reimplemented from lti::segmentation.

quadTreeSegmentation& lti::quadTreeSegmentation::operator=

(

const quadTreeSegmentation &

other

)

Alias for copy member.

Parameters:

other

the functor to be copied

Returns:

a reference to this functor object

Reimplemented from lti::functor.

bool lti::quadTreeSegmentation::split	(	const image &	img,
		const float &	threshold,
		const point &	minSize,
		tree< rectangle >::node &	qtree
	)			const [protected]

Recursively compute the quad-tree for the image.

bool lti::quadTreeSegmentation::split	(	const image &	img,
		imatrix &	mask
	)			const

Split the given image using a quad-tree strategy.

Each detected region will have its own label in the resulting mask.

Parameters:

	img	input color image
	mask	labeled mask with one integer ID per detected region.

Returns:

true if successful or false otherwise.

bool lti::quadTreeSegmentation::stats	(	const image &	img,
		const point &	from,
		const point &	to,
		trgbPixel< float > &	stats
	)			const [protected]

Compute the std.

deviations in the given region of the image

---

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

• ltiQuadTreeSegmentation.h