opentl::modalities::BackgroundSub Class Reference

GPU-based foreground/background segmentation modality. More...

Inherits opentl::modalities::Modality.

List of all members.

Public Types


Public Member Functions

 BackgroundSub (const BackgroundSub &b)
 Copy Constructor.
 BackgroundSub (opentl::modelprojection::Warp *warp, opentl::modelprojection::GLRenderer *GLRender, int camIdx)
 Modality for background subtraction and matching purpose.
virtual modalities::Modalityclone () const
 Clone the current instance.
const opentl::core::cvdata::ImagegetSegmentationImage ()
virtual void init ()
 Needed init() method for parameter handling.
virtual int matchPixLevel (const TargetPtrVector &targets, T_MEAS_PIXPtrVector &outputMeas, std::size_t partitionIdx)
 Matching on pixel level
  • compare pixel-level image data with expected map, by projecting model data under a predicted state.

virtual int preProcess (const opentl::core::cvdata::Image &image, const std::vector< std::vector< int > > &preProcessROIs)
 Preprocessing and segmentation in order to get model free unassociated data.
void updateBackgroundModel (const opentl::core::cvdata::Image &image)
 Samples from a reference image, since this is a static background segmentation Only needed for the GPU background subtractor.
virtual ~BackgroundSub ()
 Destructor.


Detailed Description

GPU-based foreground/background segmentation modality.

Member Enumeration Documentation

Enumerator:
matchUsingIntegralImage  (bool) Whether to compute and use the integral image of Z, for faster residual computation (default: false)
matchPixUsingResidualImage  (bool) Whether to compute the residual image or a scalar (= pixel count) (default: false)
useGPU  (bool) Whether to use the GPU-based background subtractor or the CPU one (default: true)
preProcessBackgroundSubGPUParam  cvprocess::BackgroundSubGPU::Param --- will be removed !
preProcessBackgroundSubParam  cvprocess::BackgroundSub::Param --- will be removed !
OFFLINE_COUNT 

Reimplemented from opentl::modalities::Modality.

Enumerator:
matchPixelRInvVariance  (double) Inverse variance of pixel-level measurement (default: 1.0)
preProcess_waitForFinished  (bool) Parameter passed on to the GPU background subtraction class (default: false)
ONLINE_COUNT 

Reimplemented from opentl::modalities::Modality.


Constructor & Destructor Documentation

opentl::modalities::BackgroundSub::BackgroundSub ( opentl::modelprojection::Warp warp,
opentl::modelprojection::GLRenderer GLRender,
int  camIdx 
)

Modality for background subtraction and matching purpose.

Parameters:
warp global instance of a warp class
GLRender global instance of a GLRenderer class
camIdx camera index
param Additional parameters for the background subtractor

opentl::modalities::BackgroundSub::BackgroundSub ( const BackgroundSub b  ) 

Copy Constructor.

Parameters:
b Source BackgroundSub instance

virtual opentl::modalities::BackgroundSub::~BackgroundSub (  )  [virtual]

Destructor.


Member Function Documentation

virtual modalities::Modality* opentl::modalities::BackgroundSub::clone (  )  const [virtual]

Clone the current instance.

Returns:
Pointer to a newly created identical instance

Implements opentl::modalities::Modality.

const opentl::core::cvdata::Image& opentl::modalities::BackgroundSub::getSegmentationImage (  ) 

virtual void opentl::modalities::BackgroundSub::init (  )  [virtual]

Needed init() method for parameter handling.

Reimplemented from opentl::modalities::Modality.

virtual int opentl::modalities::BackgroundSub::matchPixLevel ( const TargetPtrVector targets,
T_MEAS_PIXPtrVector outputMeas,
std::size_t  partitionIdx 
) [virtual]

Matching on pixel level

  • compare pixel-level image data with expected map, by projecting model data under a predicted state.

Parameters:
targets Vector of targets
outputMeas Pixel-space measurement (can be a pixel map of residuals, or a unique residual value)

Reimplemented from opentl::modalities::Modality.

virtual int opentl::modalities::BackgroundSub::preProcess ( const opentl::core::cvdata::Image image,
const std::vector< std::vector< int > > &  preProcessROIs 
) [virtual]

Preprocessing and segmentation in order to get model free unassociated data.

Parameters:
image Input sensor data (e.g. camera image in RGB)
preProcessROIs Regions of interest (x0,y0,width,height), per target

Reimplemented from opentl::modalities::Modality.

void opentl::modalities::BackgroundSub::updateBackgroundModel ( const opentl::core::cvdata::Image image  ) 

Samples from a reference image, since this is a static background segmentation Only needed for the GPU background subtractor.

Parameters:
image Reference image


Generated on Thu Jun 10 21:08:09 2010 for OpenTL by  doxygen 1.5.8