vtkAttributesErrorMetric Class Reference

#include <vtkAttributesErrorMetric.h>

Inheritance diagram for vtkAttributesErrorMetric:

Inheritance graph
[legend]
Collaboration diagram for vtkAttributesErrorMetric:

Collaboration graph
[legend]

List of all members.


Detailed Description

Objects that compute attribute-based error during cell tessellation.

It is a concrete error metric, based on an attribute criterium: the variation of the active attribute/component value from a linear ramp

See also:
vtkGenericCellTessellator vtkGenericSubdivisionErrorMetric
Tests:
vtkAttributesErrorMetric (Tests)

Definition at line 38 of file vtkAttributesErrorMetric.h.

typedef
vtkGenericSubdivisionErrorMetric 
Superclass
static int IsTypeOf (const char *type)
static vtkAttributesErrorMetricSafeDownCast (vtkObject *o)
virtual const char * GetClassName ()
virtual int IsA (const char *type)
void PrintSelf (ostream &os, vtkIndent indent)

Public Member Functions

void SetAbsoluteAttributeTolerance (double value)
void SetAttributeTolerance (double value)
virtual double GetAbsoluteAttributeTolerance ()
virtual double GetAttributeTolerance ()
int RequiresEdgeSubdivision (double *leftPoint, double *midPoint, double *rightPoint, double alpha)
double GetError (double *leftPoint, double *midPoint, double *rightPoint, double alpha)

Static Public Member Functions

static vtkAttributesErrorMetricNew ()

Protected Member Functions

 vtkAttributesErrorMetric ()
virtual ~vtkAttributesErrorMetric ()
void ComputeSquareAbsoluteAttributeTolerance ()

Protected Attributes

double AttributeTolerance
double SquareAbsoluteAttributeTolerance
double AbsoluteAttributeTolerance
int DefinedByAbsolute
vtkTimeStamp SquareAbsoluteAttributeToleranceComputeTime
double Range
vtkGenericAttributeCollectionAttributeCollection

Member Typedef Documentation

Standard VTK type and error macros.

Reimplemented from vtkGenericSubdivisionErrorMetric.

Definition at line 47 of file vtkAttributesErrorMetric.h.


Constructor & Destructor Documentation

vtkAttributesErrorMetric::vtkAttributesErrorMetric (  )  [protected]

virtual vtkAttributesErrorMetric::~vtkAttributesErrorMetric (  )  [protected, virtual]


Member Function Documentation

static vtkAttributesErrorMetric* vtkAttributesErrorMetric::New (  )  [static]

Construct the error metric with a default relative attribute accuracy equal to 0.1.

Reimplemented from vtkObject.

virtual const char* vtkAttributesErrorMetric::GetClassName (  )  [virtual]

Standard VTK type and error macros.

Reimplemented from vtkGenericSubdivisionErrorMetric.

static int vtkAttributesErrorMetric::IsTypeOf ( const char *  type  )  [static]

Standard VTK type and error macros.

Reimplemented from vtkGenericSubdivisionErrorMetric.

virtual int vtkAttributesErrorMetric::IsA ( const char *  type  )  [virtual]

Standard VTK type and error macros.

Reimplemented from vtkGenericSubdivisionErrorMetric.

static vtkAttributesErrorMetric* vtkAttributesErrorMetric::SafeDownCast ( vtkObject o  )  [static]

Standard VTK type and error macros.

Reimplemented from vtkGenericSubdivisionErrorMetric.

void vtkAttributesErrorMetric::PrintSelf ( ostream &  os,
vtkIndent  indent 
) [virtual]

Standard VTK type and error macros.

Reimplemented from vtkGenericSubdivisionErrorMetric.

virtual double vtkAttributesErrorMetric::GetAbsoluteAttributeTolerance (  )  [virtual]

Absolute tolerance of the active scalar (attribute+component). Subdivision is required if the square distance between the real attribute at the mid point on the edge and the interpolated attribute is greater than AbsoluteAttributeTolerance. This is the attribute accuracy. 0.01 will give better result than 0.1.

void vtkAttributesErrorMetric::SetAbsoluteAttributeTolerance ( double  value  ) 

Set the absolute attribute accuracy to `value'. See GetAbsoluteAttributeTolerance() for details. It is particularly useful when some concrete implementation of vtkGenericAttribute does not support GetRange() request, called internally in SetAttributeTolerance(). It may happen when the implementation support higher order attributes but cannot compute the range.

Precondition:
valid_range_value: value>0

virtual double vtkAttributesErrorMetric::GetAttributeTolerance (  )  [virtual]

Relative tolerance of the active scalar (attribute+component). Subdivision is required if the square distance between the real attribute at the mid point on the edge and the interpolated attribute is greater than AttributeTolerance. This is the attribute accuracy. 0.01 will give better result than 0.1.

void vtkAttributesErrorMetric::SetAttributeTolerance ( double  value  ) 

Set the relative attribute accuracy to `value'. See GetAttributeTolerance() for details.

Precondition:
valid_range_value: value>0 && value<1

int vtkAttributesErrorMetric::RequiresEdgeSubdivision ( double *  leftPoint,
double *  midPoint,
double *  rightPoint,
double  alpha 
) [virtual]

Does the edge need to be subdivided according to the distance between the value of the active attribute/component at the midpoint and the mean value between the endpoints? The edge is defined by its `leftPoint' and its `rightPoint'. `leftPoint', `midPoint' and `rightPoint' have to be initialized before calling RequiresEdgeSubdivision(). Their format is global coordinates, parametric coordinates and point centered attributes: xyx rst abc de... `alpha' is the normalized abscissa of the midpoint along the edge. (close to 0 means close to the left point, close to 1 means close to the right point)

Precondition:
leftPoint_exists: leftPoint!=0

midPoint_exists: midPoint!=0

rightPoint_exists: rightPoint!=0

clamped_alpha: alpha>0 && alpha<1

valid_size: sizeof(leftPoint)=sizeof(midPoint)=sizeof(rightPoint) =GetAttributeCollection()->GetNumberOfPointCenteredComponents()+6

Implements vtkGenericSubdivisionErrorMetric.

double vtkAttributesErrorMetric::GetError ( double *  leftPoint,
double *  midPoint,
double *  rightPoint,
double  alpha 
) [virtual]

Return the error at the mid-point. The type of error depends on the state of the concrete error metric. For instance, it can return an absolute or relative error metric. See RequiresEdgeSubdivision() for a description of the arguments.

Precondition:
leftPoint_exists: leftPoint!=0

midPoint_exists: midPoint!=0

rightPoint_exists: rightPoint!=0

clamped_alpha: alpha>0 && alpha<1

valid_size: sizeof(leftPoint)=sizeof(midPoint)=sizeof(rightPoint) =GetAttributeCollection()->GetNumberOfPointCenteredComponents()+6

Postcondition:
positive_result: result>=0

Implements vtkGenericSubdivisionErrorMetric.

void vtkAttributesErrorMetric::ComputeSquareAbsoluteAttributeTolerance (  )  [protected]

Compute the square absolute attribute tolerance, only if the cached value is obsolete.


Member Data Documentation

Definition at line 124 of file vtkAttributesErrorMetric.h.

Definition at line 126 of file vtkAttributesErrorMetric.h.

Definition at line 129 of file vtkAttributesErrorMetric.h.

Definition at line 130 of file vtkAttributesErrorMetric.h.

Definition at line 132 of file vtkAttributesErrorMetric.h.

double vtkAttributesErrorMetric::Range [protected]

Definition at line 134 of file vtkAttributesErrorMetric.h.

Definition at line 136 of file vtkAttributesErrorMetric.h.


The documentation for this class was generated from the following file:

Generated on Wed Jun 3 18:53:43 2009 for VTK by  doxygen 1.5.6