View Source Evision.BgSegm (Evision v0.1.26-rc0)

Link to this section Summary

Types

t()

Type that represents an BgSegm struct.

Functions

Creates a CNT Background Subtractor

Creates a CNT Background Subtractor

Creates a GMG Background Subtractor

Creates a GMG Background Subtractor

Creates an instance of BackgroundSubtractorGSOC algorithm.

Creates an instance of BackgroundSubtractorGSOC algorithm.

Creates an instance of BackgroundSubtractorLSBP algorithm.

Creates an instance of BackgroundSubtractorLSBP algorithm.

Creates mixture-of-gaussian background subtractor

Creates mixture-of-gaussian background subtractor

Creates an instance of SyntheticSequenceGenerator.

Creates an instance of SyntheticSequenceGenerator.

Link to this section Types

@type t() :: %Evision.BgSegm{ref: reference()}

Type that represents an BgSegm struct.

  • ref. reference()

    The underlying erlang resource variable.

Link to this section Functions

Link to this function

createBackgroundSubtractorCNT()

View Source
@spec createBackgroundSubtractorCNT() ::
  Evision.BgSegm.BackgroundSubtractorCNT.t() | {:error, String.t()}

Creates a CNT Background Subtractor

Keyword Arguments
  • minPixelStability: int.

    number of frames with same pixel color to consider stable

  • useHistory: bool.

    determines if we're giving a pixel credit for being stable for a long time

  • maxPixelStability: int.

    maximum allowed credit for a pixel in history

  • isParallel: bool.

    determines if we're parallelizing the algorithm

Return

Python prototype (for reference only):

createBackgroundSubtractorCNT([, minPixelStability[, useHistory[, maxPixelStability[, isParallel]]]]) -> retval
Link to this function

createBackgroundSubtractorCNT(opts)

View Source
@spec createBackgroundSubtractorCNT([{atom(), term()}, ...] | nil) ::
  Evision.BgSegm.BackgroundSubtractorCNT.t() | {:error, String.t()}

Creates a CNT Background Subtractor

Keyword Arguments
  • minPixelStability: int.

    number of frames with same pixel color to consider stable

  • useHistory: bool.

    determines if we're giving a pixel credit for being stable for a long time

  • maxPixelStability: int.

    maximum allowed credit for a pixel in history

  • isParallel: bool.

    determines if we're parallelizing the algorithm

Return

Python prototype (for reference only):

createBackgroundSubtractorCNT([, minPixelStability[, useHistory[, maxPixelStability[, isParallel]]]]) -> retval
Link to this function

createBackgroundSubtractorGMG()

View Source
@spec createBackgroundSubtractorGMG() ::
  Evision.BgSegm.BackgroundSubtractorGMG.t() | {:error, String.t()}

Creates a GMG Background Subtractor

Keyword Arguments
  • initializationFrames: int.

    number of frames used to initialize the background models.

  • decisionThreshold: double.

    Threshold value, above which it is marked foreground, else background.

Return

Python prototype (for reference only):

createBackgroundSubtractorGMG([, initializationFrames[, decisionThreshold]]) -> retval
Link to this function

createBackgroundSubtractorGMG(opts)

View Source
@spec createBackgroundSubtractorGMG([{atom(), term()}, ...] | nil) ::
  Evision.BgSegm.BackgroundSubtractorGMG.t() | {:error, String.t()}

Creates a GMG Background Subtractor

Keyword Arguments
  • initializationFrames: int.

    number of frames used to initialize the background models.

  • decisionThreshold: double.

    Threshold value, above which it is marked foreground, else background.

Return

Python prototype (for reference only):

createBackgroundSubtractorGMG([, initializationFrames[, decisionThreshold]]) -> retval
Link to this function

createBackgroundSubtractorGSOC()

View Source
@spec createBackgroundSubtractorGSOC() ::
  Evision.BgSegm.BackgroundSubtractorGSOC.t() | {:error, String.t()}

Creates an instance of BackgroundSubtractorGSOC algorithm.

Keyword Arguments
  • mc: int.

    Whether to use camera motion compensation.

  • nSamples: int.

    Number of samples to maintain at each point of the frame.

  • replaceRate: float.

    Probability of replacing the old sample - how fast the model will update itself.

  • propagationRate: float.

    Probability of propagating to neighbors.

  • hitsThreshold: int.

    How many positives the sample must get before it will be considered as a possible replacement.

  • alpha: float.

    Scale coefficient for threshold.

  • beta: float.

    Bias coefficient for threshold.

  • blinkingSupressionDecay: float.

    Blinking supression decay factor.

  • blinkingSupressionMultiplier: float.

    Blinking supression multiplier.

  • noiseRemovalThresholdFacBG: float.

    Strength of the noise removal for background points.

  • noiseRemovalThresholdFacFG: float.

    Strength of the noise removal for foreground points.

Return

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

Python prototype (for reference only):

createBackgroundSubtractorGSOC([, mc[, nSamples[, replaceRate[, propagationRate[, hitsThreshold[, alpha[, beta[, blinkingSupressionDecay[, blinkingSupressionMultiplier[, noiseRemovalThresholdFacBG[, noiseRemovalThresholdFacFG]]]]]]]]]]]) -> retval
Link to this function

createBackgroundSubtractorGSOC(opts)

View Source
@spec createBackgroundSubtractorGSOC([{atom(), term()}, ...] | nil) ::
  Evision.BgSegm.BackgroundSubtractorGSOC.t() | {:error, String.t()}

Creates an instance of BackgroundSubtractorGSOC algorithm.

Keyword Arguments
  • mc: int.

    Whether to use camera motion compensation.

  • nSamples: int.

    Number of samples to maintain at each point of the frame.

  • replaceRate: float.

    Probability of replacing the old sample - how fast the model will update itself.

  • propagationRate: float.

    Probability of propagating to neighbors.

  • hitsThreshold: int.

    How many positives the sample must get before it will be considered as a possible replacement.

  • alpha: float.

    Scale coefficient for threshold.

  • beta: float.

    Bias coefficient for threshold.

  • blinkingSupressionDecay: float.

    Blinking supression decay factor.

  • blinkingSupressionMultiplier: float.

    Blinking supression multiplier.

  • noiseRemovalThresholdFacBG: float.

    Strength of the noise removal for background points.

  • noiseRemovalThresholdFacFG: float.

    Strength of the noise removal for foreground points.

Return

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

Python prototype (for reference only):

createBackgroundSubtractorGSOC([, mc[, nSamples[, replaceRate[, propagationRate[, hitsThreshold[, alpha[, beta[, blinkingSupressionDecay[, blinkingSupressionMultiplier[, noiseRemovalThresholdFacBG[, noiseRemovalThresholdFacFG]]]]]]]]]]]) -> retval
Link to this function

createBackgroundSubtractorLSBP()

View Source
@spec createBackgroundSubtractorLSBP() ::
  Evision.BgSegm.BackgroundSubtractorLSBP.t() | {:error, String.t()}

Creates an instance of BackgroundSubtractorLSBP algorithm.

Keyword Arguments
  • mc: int.

    Whether to use camera motion compensation.

  • nSamples: int.

    Number of samples to maintain at each point of the frame.

  • lSBPRadius: int.

    LSBP descriptor radius.

  • tlower: float.

    Lower bound for T-values. See @cite LGuo2016 for details.

  • tupper: float.

    Upper bound for T-values. See @cite LGuo2016 for details.

  • tinc: float.

    Increase step for T-values. See @cite LGuo2016 for details.

  • tdec: float.

    Decrease step for T-values. See @cite LGuo2016 for details.

  • rscale: float.

    Scale coefficient for threshold values.

  • rincdec: float.

    Increase/Decrease step for threshold values.

  • noiseRemovalThresholdFacBG: float.

    Strength of the noise removal for background points.

  • noiseRemovalThresholdFacFG: float.

    Strength of the noise removal for foreground points.

  • lSBPthreshold: int.

    Threshold for LSBP binary string.

  • minCount: int.

    Minimal number of matches for sample to be considered as foreground.

Return

Background Subtraction using Local SVD Binary Pattern. More details about the algorithm can be found at @cite LGuo2016

Python prototype (for reference only):

createBackgroundSubtractorLSBP([, mc[, nSamples[, LSBPRadius[, Tlower[, Tupper[, Tinc[, Tdec[, Rscale[, Rincdec[, noiseRemovalThresholdFacBG[, noiseRemovalThresholdFacFG[, LSBPthreshold[, minCount]]]]]]]]]]]]]) -> retval
Link to this function

createBackgroundSubtractorLSBP(opts)

View Source
@spec createBackgroundSubtractorLSBP([{atom(), term()}, ...] | nil) ::
  Evision.BgSegm.BackgroundSubtractorLSBP.t() | {:error, String.t()}

Creates an instance of BackgroundSubtractorLSBP algorithm.

Keyword Arguments
  • mc: int.

    Whether to use camera motion compensation.

  • nSamples: int.

    Number of samples to maintain at each point of the frame.

  • lSBPRadius: int.

    LSBP descriptor radius.

  • tlower: float.

    Lower bound for T-values. See @cite LGuo2016 for details.

  • tupper: float.

    Upper bound for T-values. See @cite LGuo2016 for details.

  • tinc: float.

    Increase step for T-values. See @cite LGuo2016 for details.

  • tdec: float.

    Decrease step for T-values. See @cite LGuo2016 for details.

  • rscale: float.

    Scale coefficient for threshold values.

  • rincdec: float.

    Increase/Decrease step for threshold values.

  • noiseRemovalThresholdFacBG: float.

    Strength of the noise removal for background points.

  • noiseRemovalThresholdFacFG: float.

    Strength of the noise removal for foreground points.

  • lSBPthreshold: int.

    Threshold for LSBP binary string.

  • minCount: int.

    Minimal number of matches for sample to be considered as foreground.

Return

Background Subtraction using Local SVD Binary Pattern. More details about the algorithm can be found at @cite LGuo2016

Python prototype (for reference only):

createBackgroundSubtractorLSBP([, mc[, nSamples[, LSBPRadius[, Tlower[, Tupper[, Tinc[, Tdec[, Rscale[, Rincdec[, noiseRemovalThresholdFacBG[, noiseRemovalThresholdFacFG[, LSBPthreshold[, minCount]]]]]]]]]]]]]) -> retval
Link to this function

createBackgroundSubtractorMOG()

View Source
@spec createBackgroundSubtractorMOG() ::
  Evision.BgSegm.BackgroundSubtractorMOG.t() | {:error, String.t()}

Creates mixture-of-gaussian background subtractor

Keyword Arguments
  • history: int.

    Length of the history.

  • nmixtures: int.

    Number of Gaussian mixtures.

  • backgroundRatio: double.

    Background ratio.

  • noiseSigma: double.

    Noise strength (standard deviation of the brightness or each color channel). 0 means some automatic value.

Return

Python prototype (for reference only):

createBackgroundSubtractorMOG([, history[, nmixtures[, backgroundRatio[, noiseSigma]]]]) -> retval
Link to this function

createBackgroundSubtractorMOG(opts)

View Source
@spec createBackgroundSubtractorMOG([{atom(), term()}, ...] | nil) ::
  Evision.BgSegm.BackgroundSubtractorMOG.t() | {:error, String.t()}

Creates mixture-of-gaussian background subtractor

Keyword Arguments
  • history: int.

    Length of the history.

  • nmixtures: int.

    Number of Gaussian mixtures.

  • backgroundRatio: double.

    Background ratio.

  • noiseSigma: double.

    Noise strength (standard deviation of the brightness or each color channel). 0 means some automatic value.

Return

Python prototype (for reference only):

createBackgroundSubtractorMOG([, history[, nmixtures[, backgroundRatio[, noiseSigma]]]]) -> retval
Link to this function

createSyntheticSequenceGenerator(background, object)

View Source
@spec createSyntheticSequenceGenerator(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in()
) ::
  Evision.BgSegm.SyntheticSequenceGenerator.t() | {:error, String.t()}

Creates an instance of SyntheticSequenceGenerator.

Positional Arguments
  • background: Evision.Mat.

    Background image for object.

  • object: Evision.Mat.

    Object image which will move slowly over the background.

Keyword Arguments
  • amplitude: double.

    Amplitude of wave distortion applied to background.

  • wavelength: double.

    Length of waves in distortion applied to background.

  • wavespeed: double.

    How fast waves will move.

  • objspeed: double.

    How fast object will fly over background.

Return

Python prototype (for reference only):

createSyntheticSequenceGenerator(background, object[, amplitude[, wavelength[, wavespeed[, objspeed]]]]) -> retval
Link to this function

createSyntheticSequenceGenerator(background, object, opts)

View Source
@spec createSyntheticSequenceGenerator(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.BgSegm.SyntheticSequenceGenerator.t() | {:error, String.t()}

Creates an instance of SyntheticSequenceGenerator.

Positional Arguments
  • background: Evision.Mat.

    Background image for object.

  • object: Evision.Mat.

    Object image which will move slowly over the background.

Keyword Arguments
  • amplitude: double.

    Amplitude of wave distortion applied to background.

  • wavelength: double.

    Length of waves in distortion applied to background.

  • wavespeed: double.

    How fast waves will move.

  • objspeed: double.

    How fast object will fly over background.

Return

Python prototype (for reference only):

createSyntheticSequenceGenerator(background, object[, amplitude[, wavelength[, wavespeed[, objspeed]]]]) -> retval