Class BackgroundStationaryGmm_MB<T extends ImageMultiBand<T>>

All Implemented Interfaces:

public class BackgroundStationaryGmm_MB<T extends ImageMultiBand<T>>
extends BackgroundStationaryGmm<T>
Implementation of BackgroundAlgorithmGmm for ImageMultiBand.
  • Constructor Details

    • BackgroundStationaryGmm_MB

      public BackgroundStationaryGmm_MB​(float learningPeriod, float decayCoef, int maxGaussians, ImageType<T> imageType)
      learningPeriod - Specifies how fast it will adjust to changes in the image. Must be greater than zero.
      decayCoef - Determines how quickly a Gaussian is forgotten
      maxGaussians - Maximum number of Gaussians in a mixture for a pixel
      imageType - Type of image it's processing.
  • Method Details

    • updateBackground

      public void updateBackground​(T frame, @Nullable @Nullable GrayU8 mask)
      Description copied from class: BackgroundModelStationary
      Updates the background and segments it at the same time. In some implementations this can be significantly faster than doing it with separate function calls. Segmentation is performed using the model which it has prior to the update.
      updateBackground in class BackgroundStationaryGmm<T extends ImageMultiBand<T>>
      mask - If null then the background mask is ignored
    • segment

      public void segment​(T frame, GrayU8 segmented)
      Description copied from class: BackgroundModelStationary
      Invoke to segment input image into background and foreground pixels. If 'segmented' isn't the correct size it will be resized.
      Specified by:
      segment in class BackgroundModelStationary<T extends ImageMultiBand<T>>
      frame - (input) current image
      segmented - (output) Segmented image. 0 = background, 1 = foreground/moving