lti::splitImageToGSC Class Reference
[Color Analysis]

Split image in its c1, c2 and c3 components, as described in T. More...

#include <ltiSplitImageToGSC.h>

Inheritance diagram for lti::splitImageToGSC:

[legend]

Collaboration diagram for lti::splitImageToGSC:

[legend]

List of all members.

Public Member Functions
virtual const char *	getTypeName () const
virtual functor *	clone () const
virtual bool	apply (const image &img, channel &c1, channel &c2, channel &c3) const
virtual bool	apply (const image &img, channel8 &c1, channel8 &c2, channel8 &c3) const
virtual bool	apply (const rgbPixel &pixel, float &c1, float &c2, float &c3) const
virtual bool	apply (const rgbPixel &pixel, ubyte &c1, ubyte &c2, ubyte &c3) const
virtual bool	getFirst (const image &img, channel &c1) const
virtual bool	getFirst (const image &img, channel8 &c1) const
virtual bool	getSecond (const image &img, channel &c2) const
virtual bool	getSecond (const image &img, channel8 &c2) const
virtual bool	getThird (const image &img, channel &c3) const
virtual bool	getThird (const image &img, channel8 &c3) const
Protected Attributes
arctanLUT	atan2

Detailed Description

Split image in its c1, c2 and c3 components, as described in T.

Gevers and A. Smeulders "Color-based object recognition" Pattern Recognition, Vol. 32, 1999, pp. 453-464.

This color space is appropriate when trying to detect only highlight or material changes, but ignoring shadow or shape edges.

The channels are defined as follows:

$\begin{aligned} c_1 &= \arctan \left( \frac{R}{\max(G,B)} \right) \cdot \frac{2}{\pi}\\ c_2 &= \arctan \left( \frac{G}{\max(R,B)} \right) \cdot \frac{2}{\pi}\\ c_3 &= \arctan \left( \frac{B}{\max(R,G)} \right) \cdot \frac{2}{\pi} \end{aligned}$

They denote the angles of the body reflection vector and are invariant for matte, dull objects.

Please note the slightly difference with the original paper definition: all channels are normalized in order to get values between 0 and 1.

There is no merge functor for this split method.

Member Function Documentation

virtual bool lti::splitImageToGSC::apply	(	const rgbPixel &	pixel,
		ubyte &	c1,
		ubyte &	c2,
		ubyte &	c3
	)			const `[virtual]`

split the pixel in red green and blue values.

The values of each pixel will be between 0 and 255

Parameters:

	pixel	the pixel to be splitted
	c1	first channel
	c2	second channel
	c3	third channel

Implements lti::splitImage.

virtual bool lti::splitImageToGSC::apply	(	const rgbPixel &	pixel,
		float &	c1,
		float &	c2,
		float &	c3
	)			const `[virtual]`

split the pixel in red green and blue values.

The values of each pixel will be between 0.0f and 1.0f

Parameters:

	pixel	the pixel to be splitted
	c1	first channel
	c2	second channel
	c3	third channel

Implements lti::splitImage.

virtual bool lti::splitImageToGSC::apply	(	const image &	img,
		channel8 &	c1,
		channel8 &	c2,
		channel8 &	c3
	)			const `[virtual]`

split the image in red green and blue channels.

The values of each pixel will be between 0 and 255

Parameters:

	img	the image to be splitted
	c1	first channel
	c2	second channel
	c3	third channel

Implements lti::splitImage.

virtual bool lti::splitImageToGSC::apply	(	const image &	img,
		channel &	c1,
		channel &	c2,
		channel &	c3
	)			const `[virtual]`

split the image in red green and blue channels.

The values of each pixel will be between 0.0f and 1.0f

Parameters:

	img	the image to be splitted
	c1	first channel
	c2	second channel
	c3	third channel

Implements lti::splitImage.

virtual functor* lti::splitImageToGSC::clone ( ) const [virtual]

returns a pointer to a clone of the functor.

Implements lti::functor.

virtual bool lti::splitImageToGSC::getFirst	(	const image &	img,
		channel8 &	c1
	)			const `[virtual]`

Returns the first of the three channels into which the image is split.