icl::cv::SurfFeatureDetector Class Reference

ICL's *New* Generic Surf Feature detection class. More...

#include <SurfFeatureDetector.h>

List of all members.

Public Member Functions
	SurfFeatureDetector (int octaves=5, int intervals=4, int sampleStep=2, float threshold=0.00005f, const std::string &plugin="best")
	Constructor with given SURF detection parameters.
	~SurfFeatureDetector ()
const std::vector< SurfFeature > &	detect (const core::ImgBase *image)
	detects SURF features in given image
void	setReferenceImage (const core::ImgBase *image)
	detects SURF features in given image and stores them as internal reference features
const std::vector< SurfFeature > &	getReferenceFeatures () const
	returns the featuers internally stored as reference
const std::vector< SurfMatch > &	match (const core::ImgBase *image, float significance=0.65)
	detections SURF features and matches them against the given reference features
void	setOctaves (int octaves)
void	setIntervals (int intervals)
void	setSampleStep (int sampleStep)
void	setThreshold (float threshold)
Private Attributes
Data *	m_data
	hidden implementation

---

Detailed Description

ICL's *New* Generic Surf Feature detection class.

Internally, the class either uses the opensurf-based implementation or an OpenCL-based implementation based on the clsurf library. The OpenSurf backend needs OpenCV, while the clsurf-backend builds on mandatory OpenCL support.

---

Constructor & Destructor Documentation

icl::cv::SurfFeatureDetector::SurfFeatureDetector	(	int	octaves = 5,
		int	intervals = 4,
		int	sampleStep = 2,
		float	threshold = 0.00005f,
		const std::string &	plugin = "best"
	)

Constructor with given SURF detection parameters.

plugin can be either "opensurf" or "clsurf" or "best", which will prefer "clsurf" if possible

icl::cv::SurfFeatureDetector::~SurfFeatureDetector

(

)

---

Member Function Documentation

const std::vector<SurfFeature>& icl::cv::SurfFeatureDetector::detect

(

const core::ImgBase *

image

)

detects SURF features in given image

const std::vector<SurfFeature>& icl::cv::SurfFeatureDetector::getReferenceFeatures

(

)

const

returns the featuers internally stored as reference

const std::vector<SurfMatch>& icl::cv::SurfFeatureDetector::match	(	const core::ImgBase *	image,
		float	significance = 0.65
	)

detections SURF features and matches them against the given reference features

void icl::cv::SurfFeatureDetector::setIntervals

(

int

intervals

)

void icl::cv::SurfFeatureDetector::setOctaves

(

int

octaves

)

void icl::cv::SurfFeatureDetector::setReferenceImage

(

const core::ImgBase *

image

)

detects SURF features in given image and stores them as internal reference features

Please note, that the reference image is internally stored (by deep copy) When parameters are changed and the back-end must be re-initialized, the copy of the reference image is used to compute a new reference feature set based on the new parameters.

void icl::cv::SurfFeatureDetector::setSampleStep

(

int

sampleStep

)

void icl::cv::SurfFeatureDetector::setThreshold

(

float

threshold

)

---

Member Data Documentation

Data* icl::cv::SurfFeatureDetector::m_data [private]

hidden implementation

hidden data pointer

---

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

• /home/jenkins/workspace/icl-manual-trunk/icl-manual/ICLCV/src/ICLCV/SurfFeatureDetector.h