firevision::GradientClassifier Class Reference

Gradient classifier. More...

#include <>>

Inheritance diagram for firevision::GradientClassifier:

List of all members.

Public Member Functions
	GradientClassifier (std::list< ScanlineGrid * > *scanlines, Qualifier *q, unsigned int threshold, unsigned int max_size=0, bool use_rising_flank=true, bool use_falling_flank=true)
	Constructor.
virtual	~GradientClassifier ()
	Destructor.
virtual std::list< ROI > *	classify ()
	Classify image.
virtual void	set_src_buffer (unsigned char *yuv422_planar, unsigned int width, unsigned int height)
	Set source buffer.
virtual void	set_threshold (unsigned int threshold, unsigned int max_size=0)
	Threshold setter.
virtual void	set_edges (bool use_rising_edge, bool use_falling_edge)
	Edge setter.
Public Member Functions inherited from firevision::Classifier
	Classifier (const char *name)
	Constructor.
virtual	~Classifier ()
	Destructor.
virtual const char *	name () const
	Get name of classifier.

Additional Inherited Members
Protected Attributes inherited from firevision::Classifier
unsigned char *	_src
	Source buffer, encoded as YUV422_PLANAR.
unsigned int	_width
	Width in pixels of _src buffer.
unsigned int	_height
	Height in pixels of _src buffer.

---

Detailed Description

Gradient classifier.

Uses the difference of the current and the last value.

Definition at line 36 of file gradient.h.

---

Constructor & Destructor Documentation

firevision::GradientClassifier::GradientClassifier	(	std::list< ScanlineGrid * > *	scanlines,
		Qualifier *	q,
		unsigned int	threshold,
		unsigned int	max_size = 0,
		bool	use_rising_flank = true,
		bool	use_falling_flank = true
	)

Constructor.

Parameters:

scanlines	list of scanline models (Does only work with ScanlineGrid)
q	Qualifier for a single pixel (The qualifier gets deleted by this class)
threshold	minimum rise required for classification
max_size	of an object to be detected (if 0 value will be ignored)
use_rising_flank	if true the classification can start on a rising flank
use_falling_flank	if true the classification can start on a falling flank

Definition at line 51 of file gradient.cpp.

References set_edges(), and set_threshold().

firevision::GradientClassifier::~GradientClassifier ( )

virtual

Destructor.

Definition at line 72 of file gradient.cpp.

---

Member Function Documentation

std::list< ROI > * firevision::GradientClassifier::classify ( )

virtual

Classify image.

The current buffer is processed and scanned for the features the classifier has been written and initialized for. It returns a list of disjunct regions of interest.

Returns:

disjunct list of extracted regions of interest

Implements firevision::Classifier.

Definition at line 121 of file gradient.cpp.

References firevision::Qualifier::get(), firevision::Qualifier::get_buffer(), firevision::ROI::pixel_step, firevision::ROI::set_height(), firevision::ROI::set_pixel_step(), firevision::ROI::set_start(), firevision::ROI::set_width(), firevision::ROI::start, fawkes::point_t::x, and fawkes::point_t::y.

void firevision::GradientClassifier::set_edges ( bool  use_rising_edge, bool  use_falling_edge  )

virtual

Edge setter.

Parameters:

use_rising_edge	if true the classification can start on a rising edge
use_falling_edge	if true the classification can start on a falling edge

Definition at line 98 of file gradient.cpp.

Referenced by GradientClassifier().

void firevision::GradientClassifier::set_src_buffer ( unsigned char *  yuv422_planar, unsigned int  width, unsigned int  height  )

virtual

Set source buffer.

Parameters:

yuv422_planar	a YUV422 planar buffer with the source image to classify. The classifier may NOT modify the image in any way. If that is required the classifier shall make a copy of the image.
width	width of buffer in pixels
height	height of buffer in pixels

Reimplemented from firevision::Classifier.

Definition at line 112 of file gradient.cpp.

References firevision::Qualifier::set_buffer().

void firevision::GradientClassifier::set_threshold ( unsigned int  threshold, unsigned int  max_size = 0  )

virtual

Threshold setter.

Parameters:

threshold	minimum rise required for classification
max_size	of an object to be detected (if 0 value will not be set)

Definition at line 83 of file gradient.cpp.

Referenced by GradientClassifier().

---

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

• src/libs/fvclassifiers/gradient.h
• src/libs/fvclassifiers/gradient.cpp