OpenCV  3.2.0-dev
Open Source Computer Vision
Classes | Functions
Improved Background-Foreground Segmentation Methods

Classes

class  cv::bgsegm::BackgroundSubtractorCNT
 Background subtraction based on counting. More...
 
class  cv::bgsegm::BackgroundSubtractorGMG
 Background Subtractor module based on the algorithm given in [62] . More...
 
class  cv::bgsegm::BackgroundSubtractorMOG
 Gaussian Mixture-based Background/Foreground Segmentation Algorithm. More...
 

Functions

Ptr< BackgroundSubtractorCNTcv::bgsegm::createBackgroundSubtractorCNT (int minPixelStability=15, bool useHistory=true, int maxPixelStability=15 *60, bool isParallel=true)
 Creates a CNT Background Subtractor. More...
 
Ptr< BackgroundSubtractorGMGcv::bgsegm::createBackgroundSubtractorGMG (int initializationFrames=120, double decisionThreshold=0.8)
 Creates a GMG Background Subtractor. More...
 
Ptr< BackgroundSubtractorMOGcv::bgsegm::createBackgroundSubtractorMOG (int history=200, int nmixtures=5, double backgroundRatio=0.7, double noiseSigma=0)
 Creates mixture-of-gaussian background subtractor. More...
 

Detailed Description

Function Documentation

§ createBackgroundSubtractorCNT()

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

Creates a CNT Background Subtractor.

Parameters
minPixelStabilitynumber of frames with same pixel color to consider stable
useHistorydetermines if we're giving a pixel credit for being stable for a long time
maxPixelStabilitymaximum allowed credit for a pixel in history
isParalleldetermines if we're parallelizing the algorithm

§ createBackgroundSubtractorGMG()

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

Creates a GMG Background Subtractor.

Parameters
initializationFramesnumber of frames used to initialize the background models.
decisionThresholdThreshold value, above which it is marked foreground, else background.

§ createBackgroundSubtractorMOG()

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

Creates mixture-of-gaussian background subtractor.

Parameters
historyLength of the history.
nmixturesNumber of Gaussian mixtures.
backgroundRatioBackground ratio.
noiseSigmaNoise strength (standard deviation of the brightness or each color channel). 0 means some automatic value.