rdf::FeatureCollectionManager Class Reference

Manages FeatureCollections. Hence, each label (e.g. printed text) is stored here along with it's features retrieved from groundtruthed images. More...

#include <SuperPixelTrainer.h>

Public Member Functions
	FeatureCollectionManager (const cv::Mat &descriptors=cv::Mat(), const PixelSet &set=PixelSet())
bool	isEmpty () const
void	write (const QString &filePath) const
void	add (const FeatureCollection &collection)
void	merge (const FeatureCollectionManager &other)
	Merges two collection managers (e.g. from two images). If the same labels exist, features will be appended. More...
void	normalize (int minFeaturesPerClass, int maxFeaturesPerClass)
QVector< FeatureCollection >	collection () const
int	numFeatures () const
QString	toString () const
cv::Ptr< cv::ml::TrainData >	toCvTrainData (int maxSamples=-1) const
LabelManager	toLabelManager () const

Static Public Member Functions
static FeatureCollectionManager	read (const QString &filePath)

Protected Member Functions
cv::Mat	allFeatures () const
	Merges all features for training. More...
cv::Mat	allLabels () const
	Returns the labels in a OpenCV compilant format. More...

Protected Attributes
QVector< FeatureCollection >	mCollection

Detailed Description

Manages FeatureCollections. Hence, each label (e.g. printed text) is stored here along with it's features retrieved from groundtruthed images.

Constructor & Destructor Documentation

rdf::FeatureCollectionManager::FeatureCollectionManager	(	const cv::Mat &	descriptors = cv::Mat(),
		const PixelSet &	set = PixelSet()
	)

Member Function Documentation

void rdf::FeatureCollectionManager::add

(

const FeatureCollection &

collection

)

Here is the caller graph for this function:

cv::Mat rdf::FeatureCollectionManager::allFeatures ( ) const

protected

Merges all features for training.

Returns

cv::Mat rdf::FeatureCollectionManager::allLabels ( ) const

protected

Returns the labels in a OpenCV compilant format.

Returns

QVector< FeatureCollection > rdf::FeatureCollectionManager::collection

(

)

const

Here is the caller graph for this function:

bool rdf::FeatureCollectionManager::isEmpty

(

)

const

void rdf::FeatureCollectionManager::merge

(

const FeatureCollectionManager &

other

)

Merges two collection managers (e.g. from two images). If the same labels exist, features will be appended.

Parameters

other

The other manager.

Here is the caller graph for this function:

void rdf::FeatureCollectionManager::normalize	(	int	minFeaturesPerClass,
		int	maxFeaturesPerClass
	)

int rdf::FeatureCollectionManager::numFeatures

(

)

const

FeatureCollectionManager rdf::FeatureCollectionManager::read ( const QString &  filePath )

static

Here is the caller graph for this function:

cv::Ptr< cv::ml::TrainData > rdf::FeatureCollectionManager::toCvTrainData

(

int

maxSamples = -1

)

const

LabelManager rdf::FeatureCollectionManager::toLabelManager

(

)

const

QString rdf::FeatureCollectionManager::toString

(

)

const

void rdf::FeatureCollectionManager::write

(

const QString &

filePath

)

const

Here is the caller graph for this function:

Member Data Documentation

QVector<FeatureCollection> rdf::FeatureCollectionManager::mCollection

protected

---

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

• src/Module/SuperPixelTrainer.h
• src/Module/SuperPixelTrainer.cpp