ForegroundGaussian

Calculates foreground mask using Gaussians.

Syntax

IppStatus ippiForegroundGaussian_8u_C1R(const Ipp8u* pSrc, int srcStep, Ipp8u* pRef, int refStep, Ipp8u* pDst, int dstStep, IppiSize roiSize, IppFGGaussianState_8u_C1R* pState);

Parameters

pSrc: Pointer to the source image ROI.
srcStep: Distance in bytes between starts of consecutive lines in the source image.
pRef: Pointer to the ROI of the destination reference background mask image.
refStep: Distance in bytes between starts of consecutive lines in the reference background mask image.
pDst: Pointer to the ROI of the destination image.
dstStep: Distance in bytes between starts of consecutive lines in the destination image.
roiSize: Size of the source and destination image ROI in pixels.
pState: Pointer to the segmentation state structure.

Description

The function ippiForegroundGaussian is declared in the ippcv.h It operates with ROI (see Regions of Interest in Intel IPP).

This foreground/background segmentation algorithm processes the sequence of images and trains statistical models for foreground and background pixels for this sequence. The algorithm applies the model to each image, classifies its pixels as belonging to the foreground or background of the scene, and then updates the model with the image data. Figure “Foreground/Background Segmentation Using the Gaussian Model” shows the scheme of the algorithm using the Gaussian mixture statistical model for each pixel [Kadew01]. The function ippiForegroundGaussian applies the Gaussian mixture statistical model to the image pSrc, and sets pixels of pMask image corresponding to the background pixels of the pSrc to zero, and foreground pixels - to non-zero value. Then the statistical model is updated. The mask image pMask is usually processed to delete foreground and background outlier pixels. The function calculates also the reference background image pRef.

The function ippiForegroundGaussian can process images with roiSize that is not greater than roiSize specified in the function ippiForegroundGaussianInitAlloc.

Foreground/Background Segmentation Using the Gaussian Model

Return Values

ippStsNoErr	Indicates no error. Any other value indicates an error or a warning.
ippStsNullPtrErr	Indicates an error condition if one of the specified pointers is NULL.
ippStsSizeErr	Indicates an error condition if roiSize has a field with zero or negative value.
ippStsStepErr	Indicates an error condition if one of the srcStep, dstStep, or refStep is less than roiSize.width * < pixelSize>.