Improved Background-Foreground Segmentation Methods

Detailed Description

Classes
class	cv::bgsegm::BackgroundSubtractorCNT
	Background subtraction based on counting. More...
class	cv::bgsegm::BackgroundSubtractorGMG
	Background Subtractor module based on the algorithm given in [106] . More...
class	cv::bgsegm::BackgroundSubtractorGSOC
	Implementation of the different yet better algorithm which is called GSOC, as it was implemented during GSOC and was not originated from any paper. More...
class	cv::bgsegm::BackgroundSubtractorLSBP
	Background Subtraction using Local SVD Binary Pattern. More details about the algorithm can be found at [115]. More...
class	cv::bgsegm::BackgroundSubtractorLSBPDesc
	This is for calculation of the LSBP descriptors. More...
class	cv::bgsegm::BackgroundSubtractorMOG
	Gaussian Mixture-based Background/Foreground Segmentation Algorithm. More...
class	cv::bgsegm::SyntheticSequenceGenerator
	Synthetic frame sequence generator for testing background subtraction algorithms. More...

Enumerations
enum	cv::bgsegm::LSBPCameraMotionCompensation {   cv::bgsegm::LSBP_CAMERA_MOTION_COMPENSATION_NONE = 0 ,   cv::bgsegm::LSBP_CAMERA_MOTION_COMPENSATION_LK }

Functions
Ptr< BackgroundSubtractorCNT >	cv::bgsegm::createBackgroundSubtractorCNT (int minPixelStability=15, bool useHistory=true, int maxPixelStability=15 *60, bool isParallel=true)
	Creates a CNT Background Subtractor.
Ptr< BackgroundSubtractorGMG >	cv::bgsegm::createBackgroundSubtractorGMG (int initializationFrames=120, double decisionThreshold=0.8)
	Creates a GMG Background Subtractor.
Ptr< BackgroundSubtractorGSOC >	cv::bgsegm::createBackgroundSubtractorGSOC (int mc=LSBP_CAMERA_MOTION_COMPENSATION_NONE, int nSamples=20, float replaceRate=0.003f, float propagationRate=0.01f, int hitsThreshold=32, float alpha=0.01f, float beta=0.0022f, float blinkingSupressionDecay=0.1f, float blinkingSupressionMultiplier=0.1f, float noiseRemovalThresholdFacBG=0.0004f, float noiseRemovalThresholdFacFG=0.0008f)
	Creates an instance of BackgroundSubtractorGSOC algorithm.
Ptr< BackgroundSubtractorLSBP >	cv::bgsegm::createBackgroundSubtractorLSBP (int mc=LSBP_CAMERA_MOTION_COMPENSATION_NONE, int nSamples=20, int LSBPRadius=16, float Tlower=2.0f, float Tupper=32.0f, float Tinc=1.0f, float Tdec=0.05f, float Rscale=10.0f, float Rincdec=0.005f, float noiseRemovalThresholdFacBG=0.0004f, float noiseRemovalThresholdFacFG=0.0008f, int LSBPthreshold=8, int minCount=2)
	Creates an instance of BackgroundSubtractorLSBP algorithm.
Ptr< BackgroundSubtractorMOG >	cv::bgsegm::createBackgroundSubtractorMOG (int history=200, int nmixtures=5, double backgroundRatio=0.7, double noiseSigma=0)
	Creates mixture-of-gaussian background subtractor.
Ptr< SyntheticSequenceGenerator >	cv::bgsegm::createSyntheticSequenceGenerator (InputArray background, InputArray object, double amplitude=2.0, double wavelength=20.0, double wavespeed=0.2, double objspeed=6.0)
	Creates an instance of SyntheticSequenceGenerator.

Enumeration Type Documentation

LSBPCameraMotionCompensation

enum cv::bgsegm::LSBPCameraMotionCompensation

#include <opencv2/bgsegm.hpp>

Enumerator
LSBP_CAMERA_MOTION_COMPENSATION_NONE  Python: cv.bgsegm.LSBP_CAMERA_MOTION_COMPENSATION_NONE
LSBP_CAMERA_MOTION_COMPENSATION_LK  Python: cv.bgsegm.LSBP_CAMERA_MOTION_COMPENSATION_LK

Function Documentation

createBackgroundSubtractorCNT()

Ptr< BackgroundSubtractorCNT > cv::bgsegm::createBackgroundSubtractorCNT	(	int	minPixelStability = 15,
		bool	useHistory = true,
		int	maxPixelStability = 15 *60,
		bool	isParallel = true
	)

Python:
	cv.bgsegm.createBackgroundSubtractorCNT(	[, minPixelStability[, useHistory[, maxPixelStability[, isParallel]]]]	) ->	retval

#include <opencv2/bgsegm.hpp>

Creates a CNT Background Subtractor.

Parameters

minPixelStability	number of frames with same pixel color to consider stable
useHistory	determines if we're giving a pixel credit for being stable for a long time
maxPixelStability	maximum allowed credit for a pixel in history
isParallel	determines if we're parallelizing the algorithm

createBackgroundSubtractorGMG()

Ptr< BackgroundSubtractorGMG > cv::bgsegm::createBackgroundSubtractorGMG	(	int	initializationFrames = 120,
		double	decisionThreshold = 0.8
	)

Python:
	cv.bgsegm.createBackgroundSubtractorGMG(	[, initializationFrames[, decisionThreshold]]	) ->	retval

#include <opencv2/bgsegm.hpp>

Creates a GMG Background Subtractor.

Parameters

initializationFrames	number of frames used to initialize the background models.
decisionThreshold	Threshold value, above which it is marked foreground, else background.

createBackgroundSubtractorGSOC()

Ptr< BackgroundSubtractorGSOC > cv::bgsegm::createBackgroundSubtractorGSOC	(	int	mc = LSBP_CAMERA_MOTION_COMPENSATION_NONE,
		int	nSamples = 20,
		float	replaceRate = 0.003f,
		float	propagationRate = 0.01f,
		int	hitsThreshold = 32,
		float	alpha = 0.01f,
		float	beta = 0.0022f,
		float	blinkingSupressionDecay = 0.1f,
		float	blinkingSupressionMultiplier = 0.1f,
		float	noiseRemovalThresholdFacBG = 0.0004f,
		float	noiseRemovalThresholdFacFG = 0.0008f
	)

Python:
	cv.bgsegm.createBackgroundSubtractorGSOC(	[, mc[, nSamples[, replaceRate[, propagationRate[, hitsThreshold[, alpha[, beta[, blinkingSupressionDecay[, blinkingSupressionMultiplier[, noiseRemovalThresholdFacBG[, noiseRemovalThresholdFacFG]]]]]]]]]]]	) ->	retval

#include <opencv2/bgsegm.hpp>

Creates an instance of BackgroundSubtractorGSOC algorithm.

Implementation of the different yet better algorithm which is called GSOC, as it was implemented during GSOC and was not originated from any paper.

Parameters

mc	Whether to use camera motion compensation.
nSamples	Number of samples to maintain at each point of the frame.
replaceRate	Probability of replacing the old sample - how fast the model will update itself.
propagationRate	Probability of propagating to neighbors.
hitsThreshold	How many positives the sample must get before it will be considered as a possible replacement.
alpha	Scale coefficient for threshold.
beta	Bias coefficient for threshold.
blinkingSupressionDecay	Blinking supression decay factor.
blinkingSupressionMultiplier	Blinking supression multiplier.
noiseRemovalThresholdFacBG	Strength of the noise removal for background points.
noiseRemovalThresholdFacFG	Strength of the noise removal for foreground points.

Here is the call graph for this function:

createBackgroundSubtractorLSBP()

Ptr< BackgroundSubtractorLSBP > cv::bgsegm::createBackgroundSubtractorLSBP	(	int	mc = LSBP_CAMERA_MOTION_COMPENSATION_NONE,
		int	nSamples = 20,
		int	LSBPRadius = 16,
		float	Tlower = 2.0f,
		float	Tupper = 32.0f,
		float	Tinc = 1.0f,
		float	Tdec = 0.05f,
		float	Rscale = 10.0f,
		float	Rincdec = 0.005f,
		float	noiseRemovalThresholdFacBG = 0.0004f,
		float	noiseRemovalThresholdFacFG = 0.0008f,
		int	LSBPthreshold = 8,
		int	minCount = 2
	)

Python:
	cv.bgsegm.createBackgroundSubtractorLSBP(	[, mc[, nSamples[, LSBPRadius[, Tlower[, Tupper[, Tinc[, Tdec[, Rscale[, Rincdec[, noiseRemovalThresholdFacBG[, noiseRemovalThresholdFacFG[, LSBPthreshold[, minCount]]]]]]]]]]]]]	) ->	retval

#include <opencv2/bgsegm.hpp>

Creates an instance of BackgroundSubtractorLSBP algorithm.

Background Subtraction using Local SVD Binary Pattern. More details about the algorithm can be found at [115]

Parameters

mc	Whether to use camera motion compensation.
nSamples	Number of samples to maintain at each point of the frame.
LSBPRadius	LSBP descriptor radius.
Tlower	Lower bound for T-values. See [115] for details.
Tupper	Upper bound for T-values. See [115] for details.
Tinc	Increase step for T-values. See [115] for details.
Tdec	Decrease step for T-values. See [115] for details.
Rscale	Scale coefficient for threshold values.
Rincdec	Increase/Decrease step for threshold values.
noiseRemovalThresholdFacBG	Strength of the noise removal for background points.
noiseRemovalThresholdFacFG	Strength of the noise removal for foreground points.
LSBPthreshold	Threshold for LSBP binary string.
minCount	Minimal number of matches for sample to be considered as foreground.

Here is the call graph for this function:

createBackgroundSubtractorMOG()

Ptr< BackgroundSubtractorMOG > cv::bgsegm::createBackgroundSubtractorMOG	(	int	history = 200,
		int	nmixtures = 5,
		double	backgroundRatio = 0.7,
		double	noiseSigma = 0
	)

Python:
	cv.bgsegm.createBackgroundSubtractorMOG(	[, history[, nmixtures[, backgroundRatio[, noiseSigma]]]]	) ->	retval

#include <opencv2/bgsegm.hpp>

Creates mixture-of-gaussian background subtractor.

Parameters

history	Length of the history.
nmixtures	Number of Gaussian mixtures.
backgroundRatio	Background ratio.
noiseSigma	Noise strength (standard deviation of the brightness or each color channel). 0 means some automatic value.

createSyntheticSequenceGenerator()

Ptr< SyntheticSequenceGenerator > cv::bgsegm::createSyntheticSequenceGenerator	(	InputArray	background,
		InputArray	object,
		double	amplitude = 2.0,
		double	wavelength = 20.0,
		double	wavespeed = 0.2,
		double	objspeed = 6.0
	)

Python:
	cv.bgsegm.createSyntheticSequenceGenerator(	background, object[, amplitude[, wavelength[, wavespeed[, objspeed]]]]	) ->	retval

#include <opencv2/bgsegm.hpp>

Creates an instance of SyntheticSequenceGenerator.

Parameters

background	Background image for object.
object	Object image which will move slowly over the background.
amplitude	Amplitude of wave distortion applied to background.
wavelength	Length of waves in distortion applied to background.
wavespeed	How fast waves will move.
objspeed	How fast object will fly over background.

Here is the call graph for this function: