#include <itkStructuralSimilarityImageFilter.h>

Detailed Description

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>
class itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >

Computes the (multi-scale) Structural Similarity Index Measure between two images.

This filter computes the Structural Similarity Index Measure (SSIM) [150] and its multi-scale extension (MS-SSIM) [149] between two input images of identical geometry. The output image stores the per-pixel SSIM map at the original resolution. The scalar SSIM, or MS-SSIM when there is more than one scale, is available via GetMSSSIM() after Update().

Local statistics: With \(G_\sigma\) the DiscreteGaussianImageFilter kernel of standard deviation \(\sigma\) (in physical units), \(\mu_x = G_\sigma * x\), \(\sigma_x^2 = G_\sigma * x^2 - \mu_x^2\) (likewise for \(y\)) and \(\sigma_{xy} = G_\sigma * (xy) - \mu_x\mu_y\) (clamped to \(|\sigma_{xy}| \le \sigma_x\sigma_y\) against round-off), the luminance and contrast-structure terms are
\[ l = \frac{2\mu_x\mu_y + C_1}{\mu_x^2 + \mu_y^2 + C_1}, \qquad cs = \frac{2\sigma_{xy} + C_2}{\sigma_x^2 + \sigma_y^2 + C_2}, \qquad \mathrm{ssim} = l \cdot cs, \]
with \(C_1 = (K_1 L)^2\), \(C_2 = (K_2 L)^2\) and \(L\) the dynamic range. If any of LuminanceExponent ( \(\alpha\)), ContrastExponent ( \(\beta\)) or StructureExponent ( \(\gamma\)) differs from 1, \(cs = c^\beta s^\gamma\) and \(\mathrm{ssim} = l^\alpha c^\beta s^\gamma\) with the separate contrast and structure terms of [150] and \(C_3 = C_2/2\). In that case \(l\) and \(s\) are clamped at 0 before exponentiation, because a negative base with a fractional exponent is undefined; unlike the unit-exponent form, that map is non-negative.

Multi-scale combination: Scale \(j = 0, \dots, M-1\) uses the inputs downsampled \(j\) times by BinShrinkImageFilter (averaging of 2x2...2 blocks) and a window of \(\sigma_j = 2^j \sigma\), i.e. the same number of pixels at every scale. The Gaussian window only computes the local statistics of a scale; the next scale is shrunk from the unsmoothed images of the current one. Each map is averaged over the whole image at its scale, giving \(\overline{cs}_j\) and \(\overline{\mathrm{ssim}}_j\). With weights \(w_j\) (ScaleWeights),
\[ \mathrm{MS\mbox{-}SSIM}(x,y) = \max(\overline{\mathrm{ssim}}_{M-1}, \epsilon)^{w_{M-1}} \prod_{j=0}^{M-2} \max(\overline{cs}_j, \epsilon)^{w_j}, \qquad \epsilon = \mathrm{MinimumScaleValue} = 10^{-6}. \]
The floor keeps the product finite and away from zero when a per-scale mean is not positive. The default weights are WangEtAl2003ScaleWeights(). With a single scale ( \(M = 1\)) the result is the plain, unfloored mean SSIM \(\overline{\mathrm{ssim}}_0\), which can be negative, and the weight value is not used. Weights must be finite and non-negative. Each image dimension must survive \(M-1\) BinShrinkImageFilter halvings, i.e. have at least \(2^{M-1}\) pixels (more when the region's start index is not a multiple of \(2^{M-1}\)).

Eq. 7 of [149] can be read as exponentiating per-pixel terms. This filter follows the authors' reference MATLAB implementation (msssim.m) instead: each scale's map is pooled to its mean before the weights are applied, and the low-pass filter before subsampling is a 2x2 average. Results still differ slightly from msssim.m, which averages each map only over pixels whose window lies fully inside the image, while this filter averages over the whole image; msssim.m also does not floor the per-scale means at MinimumScaleValue.

The filter is N-dimensional and multi-threaded. The output pixel type defaults to float.

See also: SimilarityIndexImageFilter; DiscreteGaussianImageFilter; BinShrinkImageFilter

Definition at line 106 of file itkStructuralSimilarityImageFilter.h.

Inheritance diagram for itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >:

Collaboration diagram for itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >:

Public Types
using	ConstPointer = SmartPointer<const Self>

using	IndexType = typename InputImageType::IndexType

using	InputImageRegionType = typename InputImageType::RegionType

using	InputImageType = TInputImage

using	InputPixelType = typename InputImageType::PixelType

using	OutputImageRegionType = typename OutputImageType::RegionType

using	OutputImageType = TOutputImage

using	OutputPixelType = typename OutputImageType::PixelType

using	Pointer = SmartPointer<Self>

using	RealType = typename NumericTraits<InputPixelType>::RealType

using	ScaleValuesType = Array<RealType>

using	ScaleWeightsType = Array<RealType>

using	Self = StructuralSimilarityImageFilter

using	SizeType = typename InputImageType::SizeType

using	Superclass = ImageToImageFilter<TInputImage, TOutputImage>

Public Types inherited from itk::ImageToImageFilter< TInputImage, TOutputImage >
using	ConstPointer = SmartPointer<const Self>

using	InputImageConstPointer = typename InputImageType::ConstPointer

using	InputImagePixelType = typename InputImageType::PixelType

using	InputImagePointer = typename InputImageType::Pointer

using	InputImageRegionType = typename InputImageType::RegionType

using	InputImageType = TInputImage

using	OutputImagePixelType

using	OutputImageRegionType

using	Pointer = SmartPointer<Self>

using	Self = ImageToImageFilter

using	Superclass = ImageSource<TOutputImage>

Public Types inherited from itk::ImageSource< TOutputImage >
using	ConstPointer = SmartPointer<const Self>

using	DataObjectIdentifierType = Superclass::DataObjectIdentifierType

using	DataObjectPointer = DataObject::Pointer

using	DataObjectPointerArraySizeType = Superclass::DataObjectPointerArraySizeType

using	OutputImagePixelType = typename OutputImageType::PixelType

using	OutputImagePointer = typename OutputImageType::Pointer

using	OutputImageRegionType = typename OutputImageType::RegionType

using	OutputImageType = TOutputImage

using	Pointer = SmartPointer<Self>

using	Self = ImageSource

using	Superclass = ProcessObject

Public Types inherited from itk::ProcessObject
using	ConstPointer = SmartPointer<const Self>

using	DataObjectIdentifierType = DataObject::DataObjectIdentifierType

using	DataObjectPointer = DataObject::Pointer

using	DataObjectPointerArray = std::vector<DataObjectPointer>

using	DataObjectPointerArraySizeType = DataObjectPointerArray::size_type

using	NameArray = std::vector<DataObjectIdentifierType>

using	Pointer = SmartPointer<Self>

using	Self = ProcessObject

using	Superclass = Object

Public Types inherited from itk::Object
using	ConstPointer = SmartPointer<const Self>

using	Pointer = SmartPointer<Self>

using	Self = Object

using	Superclass = LightObject

Public Types inherited from itk::LightObject
using	ConstPointer = SmartPointer<const Self>

using	Pointer = SmartPointer<Self>

using	Self = LightObject

Public Member Functions
virtual::itk::LightObject::Pointer	CreateAnother () const

virtual double	GetMSSSIM () const

virtual const char *	GetNameOfClass () const


void	SetInput1 (const InputImageType *image)

const InputImageType *	GetInput1 () const


void	SetInput2 (const InputImageType *image)

const InputImageType *	GetInput2 () const


virtual void	SetGaussianSigma (double _arg)

virtual double	GetGaussianSigma () const


virtual void	SetK1 (double _arg)

virtual double	GetK1 () const


virtual void	SetK2 (double _arg)

virtual double	GetK2 () const


virtual void	SetDynamicRange (double _arg)

virtual double	GetDynamicRange () const


virtual void	SetLuminanceExponent (double _arg)

virtual double	GetLuminanceExponent () const


virtual void	SetContrastExponent (double _arg)

virtual double	GetContrastExponent () const


virtual void	SetStructureExponent (double _arg)

virtual double	GetStructureExponent () const


void	SetScaleWeights (const ScaleWeightsType &weights)

virtual const ScaleWeightsType &	GetScaleWeights () const


virtual const ScaleValuesType &	GetSSIMPerScale () const

virtual const ScaleValuesType &	GetContrastStructurePerScale () const

Public Member Functions inherited from itk::ImageToImageFilter< TInputImage, TOutputImage >
const InputImageType *	GetInput () const

const InputImageType *	GetInput (unsigned int idx) const

void	PopBackInput () override

void	PopFrontInput () override

virtual void	PushBackInput (const InputImageType *input)

virtual void	PushFrontInput (const InputImageType *input)

virtual void	SetInput (const InputImageType *input)

virtual void	SetInput (unsigned int, const TInputImage *image)

virtual void	SetCoordinateTolerance (double _arg)

virtual double	GetCoordinateTolerance () const

virtual void	SetDirectionTolerance (double _arg)

virtual double	GetDirectionTolerance () const

Public Member Functions inherited from itk::ImageSource< TOutputImage >
OutputImageType *	GetOutput (unsigned int idx)

virtual void	GraftNthOutput (unsigned int idx, DataObject *graft)

virtual void	GraftOutput (const DataObjectIdentifierType &key, DataObject *graft)

virtual void	GraftOutput (DataObject *graft)

OutputImageType *	GetOutput ()

const OutputImageType *	GetOutput () const

ProcessObject::DataObjectPointer	MakeOutput (ProcessObject::DataObjectPointerArraySizeType idx) override

ProcessObject::DataObjectPointer	MakeOutput (const ProcessObject::DataObjectIdentifierType &) override

Public Member Functions inherited from itk::ProcessObject
virtual void	AbortGenerateDataOff ()

virtual void	AbortGenerateDataOn ()

virtual const bool &	GetAbortGenerateData () const

DataObjectPointerArray	GetIndexedInputs ()

DataObjectPointerArray	GetIndexedOutputs ()

NameArray	GetInputNames () const

DataObjectPointerArray	GetInputs ()

MultiThreaderBase *	GetMultiThreader () const

DataObjectPointerArraySizeType	GetNumberOfIndexedInputs () const

DataObjectPointerArraySizeType	GetNumberOfIndexedOutputs () const

DataObjectPointerArraySizeType	GetNumberOfInputs () const

DataObjectPointerArraySizeType	GetNumberOfOutputs () const

virtual DataObjectPointerArraySizeType	GetNumberOfValidRequiredInputs () const

NameArray	GetOutputNames () const

DataObjectPointerArray	GetOutputs ()

virtual float	GetProgress () const

NameArray	GetRequiredInputNames () const

bool	HasInput (const DataObjectIdentifierType &key) const

bool	HasOutput (const DataObjectIdentifierType &key) const

void	IncrementProgress (float increment)

virtual void	PrepareOutputs ()

virtual void	PropagateRequestedRegion (DataObject *output)

virtual void	ResetPipeline ()

virtual void	SetAbortGenerateData (bool _arg)

void	SetMultiThreader (MultiThreaderBase *threader)

virtual void	Update ()

virtual void	UpdateLargestPossibleRegion ()

virtual void	UpdateOutputData (DataObject *output)

virtual void	UpdateOutputInformation ()

void	UpdateProgress (float progress)

virtual void	SetReleaseDataFlag (bool val)

virtual bool	GetReleaseDataFlag () const

void	ReleaseDataFlagOn ()

void	ReleaseDataFlagOff ()

virtual void	SetReleaseDataBeforeUpdateFlag (bool _arg)

virtual const bool &	GetReleaseDataBeforeUpdateFlag () const

virtual void	ReleaseDataBeforeUpdateFlagOn ()

virtual void	ReleaseDataBeforeUpdateFlagOff ()

	itkVirtualSetClampMacro (NumberOfWorkUnits, ThreadIdType, 1, ITK_MAX_THREADS)

virtual const ThreadIdType &	GetNumberOfWorkUnits () const

Public Member Functions inherited from itk::Object
unsigned long	AddObserver (const EventObject &event, Command *cmd) const

unsigned long	AddObserver (const EventObject &event, std::function< void(const EventObject &)> function) const

virtual void	DebugOff () const

virtual void	DebugOn () const

Command *	GetCommand (unsigned long tag)

bool	GetDebug () const

MetaDataDictionary &	GetMetaDataDictionary ()

const MetaDataDictionary &	GetMetaDataDictionary () const

virtual ModifiedTimeType	GetMTime () const

virtual const TimeStamp &	GetTimeStamp () const

bool	HasObserver (const EventObject &event) const

void	InvokeEvent (const EventObject &)

void	InvokeEvent (const EventObject &) const

virtual void	Modified () const

void	Register () const override

void	RemoveAllObservers ()

void	RemoveObserver (unsigned long tag) const

void	SetDebug (bool debugFlag) const

void	SetReferenceCount (int) override

void	UnRegister () const noexcept override

void	SetMetaDataDictionary (const MetaDataDictionary &rhs)

void	SetMetaDataDictionary (MetaDataDictionary &&rrhs)

virtual void	SetObjectName (std::string _arg)

virtual const std::string &	GetObjectName () const

Public Member Functions inherited from itk::LightObject
Pointer	Clone () const

virtual void	Delete ()

virtual int	GetReferenceCount () const

void	Print (std::ostream &os, Indent indent=0) const

Static Public Member Functions
static ScaleWeightsType	GaussianScaleWeights (unsigned int size, double sigma)

static Pointer	New ()

static ScaleWeightsType	WangEtAl2003ScaleWeights ()

Static Public Member Functions inherited from itk::ImageToImageFilter< TInputImage, TOutputImage >
static double	GetGlobalDefaultCoordinateTolerance ()

static double	GetGlobalDefaultDirectionTolerance ()

static void	SetGlobalDefaultCoordinateTolerance (double)

static void	SetGlobalDefaultDirectionTolerance (double)

Static Public Member Functions inherited from itk::Object
static bool	GetGlobalWarningDisplay ()

static void	GlobalWarningDisplayOff ()

static void	GlobalWarningDisplayOn ()

static Pointer	New ()

static void	SetGlobalWarningDisplay (bool val)

Static Public Member Functions inherited from itk::LightObject
static void	BreakOnError ()

static Pointer	New ()

Static Public Attributes
static constexpr unsigned int	ImageDimension = InputImageType::ImageDimension

static constexpr RealType	MinimumScaleValue { 1e-6 }

Static Public Attributes inherited from itk::ImageToImageFilter< TInputImage, TOutputImage >
static constexpr unsigned int	InputImageDimension = TInputImage::ImageDimension

static constexpr unsigned int	OutputImageDimension = TOutputImage::ImageDimension

Static Public Attributes inherited from itk::ImageSource< TOutputImage >
static constexpr unsigned int	OutputImageDimension = TOutputImage::ImageDimension

Protected Member Functions
void	EnlargeOutputRequestedRegion (DataObject *data) override

void	GenerateData () override

void	PrintSelf (std::ostream &os, Indent indent) const override

	StructuralSimilarityImageFilter ()

void	VerifyPreconditions () const override

	~StructuralSimilarityImageFilter () override=default

Protected Member Functions inherited from itk::ImageToImageFilter< TInputImage, TOutputImage >
virtual void	CallCopyInputRegionToOutputRegion (OutputImageRegionType &destRegion, const InputImageRegionType &srcRegion)

virtual void	CallCopyOutputRegionToInputRegion (InputImageRegionType &destRegion, const OutputImageRegionType &srcRegion)

void	GenerateInputRequestedRegion () override

	ImageToImageFilter ()

void	VerifyInputInformation () const override

	~ImageToImageFilter () override=default

Protected Member Functions inherited from itk::ImageSource< TOutputImage >
virtual void	AfterThreadedGenerateData ()

virtual void	AllocateOutputs ()

virtual void	BeforeThreadedGenerateData ()

void	ClassicMultiThread (ThreadFunctionType callbackFunction)

virtual const ImageRegionSplitterBase *	GetImageRegionSplitter () const

	ImageSource ()

virtual unsigned int	SplitRequestedRegion (unsigned int i, unsigned int pieces, OutputImageRegionType &splitRegion)

	~ImageSource () override=default

virtual void	ThreadedGenerateData (const OutputImageRegionType &region, ThreadIdType threadId)

virtual void	DynamicThreadedGenerateData (const OutputImageRegionType &outputRegionForThread)

virtual bool	GetDynamicMultiThreading () const

virtual void	SetDynamicMultiThreading (bool _arg)

virtual void	DynamicMultiThreadingOn ()

virtual void	DynamicMultiThreadingOff ()

Protected Member Functions inherited from itk::ProcessObject
virtual void	AddInput (DataObject *input)

void	AddOptionalInputName (const DataObjectIdentifierType &)

void	AddOptionalInputName (const DataObjectIdentifierType &, DataObjectPointerArraySizeType idx)

virtual void	AddOutput (DataObject *output)

bool	AddRequiredInputName (const DataObjectIdentifierType &)

bool	AddRequiredInputName (const DataObjectIdentifierType &, DataObjectPointerArraySizeType idx)

virtual void	CacheInputReleaseDataFlags ()

virtual void	GenerateOutputInformation ()

virtual void	GenerateOutputRequestedRegion (DataObject *output)

DataObject *	GetInput (const DataObjectIdentifierType &key)

const DataObject *	GetInput (const DataObjectIdentifierType &key) const

virtual const DataObjectPointerArraySizeType &	GetNumberOfRequiredInputs () const

virtual const DataObjectPointerArraySizeType &	GetNumberOfRequiredOutputs () const

bool	IsIndexedInputName (const DataObjectIdentifierType &) const

bool	IsIndexedOutputName (const DataObjectIdentifierType &) const

bool	IsRequiredInputName (const DataObjectIdentifierType &) const

DataObjectPointerArraySizeType	MakeIndexFromInputName (const DataObjectIdentifierType &name) const

DataObjectPointerArraySizeType	MakeIndexFromOutputName (const DataObjectIdentifierType &name) const

DataObjectIdentifierType	MakeNameFromInputIndex (DataObjectPointerArraySizeType idx) const

DataObjectIdentifierType	MakeNameFromOutputIndex (DataObjectPointerArraySizeType idx) const

	ProcessObject ()

virtual void	PropagateResetPipeline ()

virtual void	PushBackInput (const DataObject *input)

virtual void	PushFrontInput (const DataObject *input)

virtual void	ReleaseInputs ()

virtual void	RemoveInput (const DataObjectIdentifierType &key)

virtual void	RemoveInput (DataObjectPointerArraySizeType)

virtual void	RemoveOutput (const DataObjectIdentifierType &key)

virtual void	RemoveOutput (DataObjectPointerArraySizeType idx)

bool	RemoveRequiredInputName (const DataObjectIdentifierType &)

virtual void	RestoreInputReleaseDataFlags ()

virtual void	SetInput (const DataObjectIdentifierType &key, DataObject *input)

virtual void	SetNthInput (DataObjectPointerArraySizeType idx, DataObject *input)

virtual void	SetNthOutput (DataObjectPointerArraySizeType idx, DataObject *output)

void	SetNumberOfIndexedInputs (DataObjectPointerArraySizeType num)

void	SetNumberOfIndexedOutputs (DataObjectPointerArraySizeType num)

virtual void	SetNumberOfRequiredInputs (DataObjectPointerArraySizeType)

virtual void	SetNumberOfRequiredOutputs (DataObjectPointerArraySizeType _arg)

virtual void	SetOutput (const DataObjectIdentifierType &name, DataObject *output)

virtual void	SetPrimaryInput (DataObject *object)

virtual void	SetPrimaryOutput (DataObject *object)

void	SetRequiredInputNames (const NameArray &)

	~ProcessObject () override

DataObject *	GetInput (DataObjectPointerArraySizeType idx)

const DataObject *	GetInput (DataObjectPointerArraySizeType idx) const

DataObject *	GetPrimaryInput ()

const DataObject *	GetPrimaryInput () const

virtual void	SetPrimaryInputName (const DataObjectIdentifierType &key)

virtual const char *	GetPrimaryInputName () const

DataObject *	GetOutput (const DataObjectIdentifierType &key)

const DataObject *	GetOutput (const DataObjectIdentifierType &key) const

virtual void	SetPrimaryOutputName (const DataObjectIdentifierType &key)

virtual const char *	GetPrimaryOutputName () const

DataObject *	GetOutput (DataObjectPointerArraySizeType i)

const DataObject *	GetOutput (DataObjectPointerArraySizeType i) const

DataObject *	GetPrimaryOutput ()

const DataObject *	GetPrimaryOutput () const

virtual bool	GetThreaderUpdateProgress () const

virtual void	ThreaderUpdateProgressOn ()

virtual void	ThreaderUpdateProgressOff ()

virtual void	SetThreaderUpdateProgress (bool arg)

Protected Member Functions inherited from itk::Object
	Object ()

bool	PrintObservers (std::ostream &os, Indent indent) const

virtual void	SetTimeStamp (const TimeStamp &timeStamp)

	~Object () override

Protected Member Functions inherited from itk::LightObject
virtual LightObject::Pointer	InternalClone () const

	LightObject ()

virtual void	PrintHeader (std::ostream &os, Indent indent) const

virtual void	PrintTrailer (std::ostream &os, Indent indent) const

virtual	~LightObject ()

Static Private Member Functions
static constexpr double	DefaultDynamicRange ()

Private Attributes
double	m_ContrastExponent { 1.0 }

ScaleValuesType	m_ContrastStructurePerScale {}

double	m_DynamicRange { DefaultDynamicRange() }

double	m_GaussianSigma { 1.5 }

double	m_K1 { 0.01 }

double	m_K2 { 0.03 }

double	m_LuminanceExponent { 1.0 }

double	m_MSSSIM { 0.0 }

ScaleWeightsType	m_ScaleWeights { WangEtAl2003ScaleWeights() }

ScaleValuesType	m_SSIMPerScale {}

double	m_StructureExponent { 1.0 }

Additional Inherited Members
Protected Types inherited from itk::ImageToImageFilter< TInputImage, TOutputImage >
using	InputToOutputRegionCopierType

using	OutputToInputRegionCopierType

Static Protected Member Functions inherited from itk::ImageSource< TOutputImage >
static const ImageRegionSplitterBase *	GetGlobalDefaultSplitter ()

static ITK_THREAD_RETURN_FUNCTION_CALL_CONVENTION	ThreaderCallback (void *arg)

Static Protected Member Functions inherited from itk::ProcessObject
template<typename TSourceObject>
static void	MakeRequiredOutputs (TSourceObject &sourceObject, const DataObjectPointerArraySizeType numberOfRequiredOutputs)

static constexpr float	progressFixedToFloat (uint32_t fixed)

static uint32_t	progressFloatToFixed (float f)

Protected Attributes inherited from itk::ImageSource< TOutputImage >
bool	m_DynamicMultiThreading { true }

Protected Attributes inherited from itk::ProcessObject
TimeStamp	m_OutputInformationMTime {}

bool	m_Updating {}

Protected Attributes inherited from itk::LightObject
std::atomic< int >	m_ReferenceCount {}

Member Typedef Documentation

◆ ConstPointer

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::ConstPointer = SmartPointer<const Self>

Definition at line 115 of file itkStructuralSimilarityImageFilter.h.

◆ IndexType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::IndexType = typename InputImageType::IndexType

Definition at line 131 of file itkStructuralSimilarityImageFilter.h.

◆ InputImageRegionType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::InputImageRegionType = typename InputImageType::RegionType

Definition at line 128 of file itkStructuralSimilarityImageFilter.h.

◆ InputImageType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::InputImageType = TInputImage

Image type aliases.

Definition at line 124 of file itkStructuralSimilarityImageFilter.h.

◆ InputPixelType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::InputPixelType = typename InputImageType::PixelType

Definition at line 126 of file itkStructuralSimilarityImageFilter.h.

◆ OutputImageRegionType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::OutputImageRegionType = typename OutputImageType::RegionType

Definition at line 129 of file itkStructuralSimilarityImageFilter.h.

◆ OutputImageType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::OutputImageType = TOutputImage

Definition at line 125 of file itkStructuralSimilarityImageFilter.h.

◆ OutputPixelType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::OutputPixelType = typename OutputImageType::PixelType

Definition at line 127 of file itkStructuralSimilarityImageFilter.h.

◆ Pointer

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::Pointer = SmartPointer<Self>

Definition at line 114 of file itkStructuralSimilarityImageFilter.h.

◆ RealType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::RealType = typename NumericTraits<InputPixelType>::RealType

Floating-point type used for all SSIM computations.

Definition at line 136 of file itkStructuralSimilarityImageFilter.h.

◆ ScaleValuesType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::ScaleValuesType = Array<RealType>

Definition at line 140 of file itkStructuralSimilarityImageFilter.h.

◆ ScaleWeightsType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::ScaleWeightsType = Array<RealType>

Type used for the multi-scale weights and the per-scale results.

Definition at line 139 of file itkStructuralSimilarityImageFilter.h.

◆ Self

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::Self = StructuralSimilarityImageFilter

Standard class type aliases.

Definition at line 112 of file itkStructuralSimilarityImageFilter.h.

◆ SizeType

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SizeType = typename InputImageType::SizeType

Definition at line 130 of file itkStructuralSimilarityImageFilter.h.

◆ Superclass

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

using itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::Superclass = ImageToImageFilter<TInputImage, TOutputImage>

Definition at line 113 of file itkStructuralSimilarityImageFilter.h.

Constructor & Destructor Documentation

◆ StructuralSimilarityImageFilter()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::StructuralSimilarityImageFilter ( )

protected

Referenced by GetNameOfClass().

◆ ~StructuralSimilarityImageFilter()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::~StructuralSimilarityImageFilter ( )

overrideprotecteddefault

Member Function Documentation

◆ CreateAnother()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual::itk::LightObject::Pointer itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::CreateAnother ( ) const

virtual

Create an object from an instance, potentially deferring to a factory. This method allows you to create an instance of an object that is exactly the same type as the referring object. This is useful in cases where an object has been cast back to a base class.

Reimplemented from itk::Object.

◆ DefaultDynamicRange()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

static constexpr double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::DefaultDynamicRange ( )

inlinestaticconstexprprivate

Default dynamic range: 1.0 for floating-point pixels (assume normalized data), and NumericTraits::max() - NumericTraits::min() for integer pixels (e.g. 255 for unsigned char).

Definition at line 278 of file itkStructuralSimilarityImageFilter.h.

References itk::NumericTraits< T >::max(), and itk::NumericTraits< T >::min().

◆ EnlargeOutputRequestedRegion()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::EnlargeOutputRequestedRegion ( DataObject * data )

overrideprotectedvirtual

This filter computes the full output.

Reimplemented from itk::ProcessObject.

◆ GaussianScaleWeights()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

static ScaleWeightsType itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GaussianScaleWeights	(	unsigned int	size,
		double	sigma )

static

Returns size samples of a Gaussian of standard deviation sigma, centered on the middle element and normalized to sum to 1. Throws if sigma is not finite and strictly positive.

◆ GenerateData()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GenerateData ( )

overrideprotectedvirtual

A version of GenerateData() specific for image processing filters. This implementation will split the processing across multiple threads. The buffer is allocated by this method. Then the BeforeThreadedGenerateData() method is called (if provided). Then, a series of threads are spawned each calling DynamicThreadedGenerateData(). After all the threads have completed processing, the AfterThreadedGenerateData() method is called (if provided). If an image processing filter cannot be threaded, the filter should provide an implementation of GenerateData(). That implementation is responsible for allocating the output buffer. If a filter can be threaded, it should NOT provide a GenerateData() method but should provide a DynamicThreadedGenerateData() instead.

See also: ThreadedGenerateData()

Reimplemented from itk::ImageSource< TOutputImage >.

◆ GetContrastExponent()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetContrastExponent ( ) const

virtual

Exponent \(\beta\) on the contrast term. Default 1.0.

◆ GetContrastStructurePerScale()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual const ScaleValuesType & itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetContrastStructurePerScale ( ) const

virtual

Per-scale means \(\overline{\mathrm{ssim}}_j\) and \(\overline{cs}_j\), finest scale first, before flooring. Available after Update().

◆ GetDynamicRange()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetDynamicRange ( ) const

virtual

Dynamic range \(L\) of the pixel values used to compute \(C_1 = (K_1 L)^2\) and \(C_2 = (K_2 L)^2\). Default depends on the input pixel type: 255 for unsigned char, 65535 for unsigned short, 1.0 for float / double, etc.

◆ GetGaussianSigma()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetGaussianSigma ( ) const

virtual

Standard deviation \(\sigma\) of the Gaussian window at the finest scale, in physical units. Default 1.5.

◆ GetInput1()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

const InputImageType * itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetInput1 ( ) const

inline

Set/Get the first input image.

Definition at line 150 of file itkStructuralSimilarityImageFilter.h.

References itk::ImageToImageFilter< TInputImage, TOutputImage >::GetInput().

◆ GetInput2()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

const InputImageType * itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetInput2 ( ) const

Set/Get the second input image.

◆ GetK1()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetK1 ( ) const

virtual

\(K_1\) stability constant. Default 0.01.

◆ GetK2()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetK2 ( ) const

virtual

\(K_2\) stability constant. Default 0.03.

◆ GetLuminanceExponent()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetLuminanceExponent ( ) const

virtual

Exponent \(\alpha\) on the luminance term. Default 1.0.

◆ GetMSSSIM()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetMSSSIM ( ) const

virtual

(MS-)SSIM. Available after Update(). With one scale it is the spatial mean of the output SSIM map, \(\overline{\mathrm{ssim}}_0\). With more scales it is not a mean but the MS-SSIM product of floored per-scale means given in the class documentation.

◆ GetNameOfClass()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual const char * itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetNameOfClass ( ) const

virtual

See also: LightObject::GetNameOfClass()

Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >.

References StructuralSimilarityImageFilter().

◆ GetScaleWeights()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual const ScaleWeightsType & itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetScaleWeights ( ) const

virtual

Per-scale exponents \(w_j\); the number of elements is the number of scales. A single element requests single-scale SSIM. Default WangEtAl2003ScaleWeights().

◆ GetSSIMPerScale()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual const ScaleValuesType & itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetSSIMPerScale ( ) const

virtual

Per-scale means \(\overline{\mathrm{ssim}}_j\) and \(\overline{cs}_j\), finest scale first, before flooring. Available after Update().

◆ GetStructureExponent()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::GetStructureExponent ( ) const

virtual

Exponent \(\gamma\) on the structure term. Default 1.0.

◆ New()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

static Pointer itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::New ( )

static

Method for creation through the object factory.

◆ PrintSelf()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::PrintSelf	(	std::ostream &	os,
		Indent	indent ) const

overrideprotectedvirtual

Methods invoked by Print() to print information about the object including superclasses. Typically not called by the user (use Print() instead) but used in the hierarchical print process to combine the output of several classes.

Reimplemented from itk::ImageToImageFilter< TInputImage, TOutputImage >.

◆ SetContrastExponent()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetContrastExponent ( double _arg )

virtual

Exponent \(\beta\) on the contrast term. Default 1.0.

◆ SetDynamicRange()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetDynamicRange ( double _arg )

virtual

Dynamic range \(L\) of the pixel values used to compute \(C_1 = (K_1 L)^2\) and \(C_2 = (K_2 L)^2\). Default depends on the input pixel type: 255 for unsigned char, 65535 for unsigned short, 1.0 for float / double, etc.

◆ SetGaussianSigma()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetGaussianSigma ( double _arg )

virtual

Standard deviation \(\sigma\) of the Gaussian window at the finest scale, in physical units. Default 1.5.

◆ SetInput1()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetInput1 ( const InputImageType * image )

inline

Set/Get the first input image.

Definition at line 145 of file itkStructuralSimilarityImageFilter.h.

References itk::ImageToImageFilter< TInputImage, TOutputImage >::SetInput().

◆ SetInput2()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetInput2 ( const InputImageType * image )

Set/Get the second input image.

◆ SetK1()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetK1 ( double _arg )

virtual

\(K_1\) stability constant. Default 0.01.

◆ SetK2()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetK2 ( double _arg )

virtual

\(K_2\) stability constant. Default 0.03.

◆ SetLuminanceExponent()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetLuminanceExponent ( double _arg )

virtual

Exponent \(\alpha\) on the luminance term. Default 1.0.

◆ SetScaleWeights()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetScaleWeights ( const ScaleWeightsType & weights )

Per-scale exponents \(w_j\); the number of elements is the number of scales. A single element requests single-scale SSIM. Default WangEtAl2003ScaleWeights().

◆ SetStructureExponent()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

virtual void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::SetStructureExponent ( double _arg )

virtual

Exponent \(\gamma\) on the structure term. Default 1.0.

◆ VerifyPreconditions()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

void itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::VerifyPreconditions ( ) const

overrideprotectedvirtual

Verify that parameters are valid and both inputs are set with matching regions.

Reimplemented from itk::ProcessObject.

◆ WangEtAl2003ScaleWeights()

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

static ScaleWeightsType itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::WangEtAl2003ScaleWeights ( )

static

Returns the 5-scale MS-SSIM weights of Wang et al. 2003 [149] \((0.0448,\ 0.2856,\ 0.3001,\ 0.2363,\ 0.1333)\).

Member Data Documentation

◆ ImageDimension

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

unsigned int itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::ImageDimension = InputImageType::ImageDimension

staticconstexpr

Definition at line 133 of file itkStructuralSimilarityImageFilter.h.

◆ m_ContrastExponent

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_ContrastExponent { 1.0 }

private

Definition at line 292 of file itkStructuralSimilarityImageFilter.h.

◆ m_ContrastStructurePerScale

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

ScaleValuesType itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_ContrastStructurePerScale {}

private

Definition at line 299 of file itkStructuralSimilarityImageFilter.h.

◆ m_DynamicRange

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_DynamicRange { DefaultDynamicRange() }

private

Definition at line 290 of file itkStructuralSimilarityImageFilter.h.

◆ m_GaussianSigma

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_GaussianSigma { 1.5 }

private

Definition at line 270 of file itkStructuralSimilarityImageFilter.h.

◆ m_K1

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_K1 { 0.01 }

private

Definition at line 271 of file itkStructuralSimilarityImageFilter.h.

◆ m_K2

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_K2 { 0.03 }

private

Definition at line 272 of file itkStructuralSimilarityImageFilter.h.

◆ m_LuminanceExponent

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_LuminanceExponent { 1.0 }

private

Definition at line 291 of file itkStructuralSimilarityImageFilter.h.

◆ m_MSSSIM

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_MSSSIM { 0.0 }

private

Definition at line 297 of file itkStructuralSimilarityImageFilter.h.

◆ m_ScaleWeights

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

ScaleWeightsType itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_ScaleWeights { WangEtAl2003ScaleWeights() }

private

Definition at line 295 of file itkStructuralSimilarityImageFilter.h.

◆ m_SSIMPerScale

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

ScaleValuesType itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_SSIMPerScale {}

private

Definition at line 298 of file itkStructuralSimilarityImageFilter.h.

◆ m_StructureExponent

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

double itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::m_StructureExponent { 1.0 }

private

Definition at line 293 of file itkStructuralSimilarityImageFilter.h.

◆ MinimumScaleValue

template<typename TInputImage, typename TOutputImage = Image<float, TInputImage::ImageDimension>>

RealType itk::StructuralSimilarityImageFilter< TInputImage, TOutputImage >::MinimumScaleValue { 1e-6 }

staticconstexpr

Floor \(\epsilon\) applied to the per-scale means in the MS-SSIM product, keeping it finite and non-zero when a mean is not positive.

Definition at line 233 of file itkStructuralSimilarityImageFilter.h.

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

itkStructuralSimilarityImageFilter.h

Detailed Description

Public Types

Public Member Functions

Static Public Member Functions

Static Public Attributes

Protected Member Functions

Static Private Member Functions

Private Attributes

Additional Inherited Members

Member Typedef Documentation

◆ ConstPointer

◆ IndexType

◆ InputImageRegionType

◆ InputImageType

◆ InputPixelType

◆ OutputImageRegionType

◆ OutputImageType

◆ OutputPixelType

◆ Pointer

◆ RealType

◆ ScaleValuesType

◆ ScaleWeightsType

◆ Self

◆ SizeType

◆ Superclass

Constructor & Destructor Documentation

◆ StructuralSimilarityImageFilter()

◆ ~StructuralSimilarityImageFilter()

Member Function Documentation

◆ CreateAnother()

◆ DefaultDynamicRange()

◆ EnlargeOutputRequestedRegion()

◆ GaussianScaleWeights()

◆ GenerateData()

◆ GetContrastExponent()

◆ GetContrastStructurePerScale()

◆ GetDynamicRange()

◆ GetGaussianSigma()

◆ GetInput1()

◆ GetInput2()

◆ GetK1()

◆ GetK2()

◆ GetLuminanceExponent()

◆ GetMSSSIM()

◆ GetNameOfClass()

◆ GetScaleWeights()

◆ GetSSIMPerScale()

◆ GetStructureExponent()

◆ New()

◆ PrintSelf()

◆ SetContrastExponent()

◆ SetDynamicRange()

◆ SetGaussianSigma()

◆ SetInput1()

◆ SetInput2()

◆ SetK1()

◆ SetK2()

◆ SetLuminanceExponent()

◆ SetScaleWeights()

◆ SetStructureExponent()

◆ VerifyPreconditions()

◆ WangEtAl2003ScaleWeights()

Member Data Documentation

◆ ImageDimension

◆ m_ContrastExponent

◆ m_ContrastStructurePerScale

◆ m_DynamicRange

◆ m_GaussianSigma

◆ m_K1

◆ m_K2

◆ m_LuminanceExponent

◆ m_MSSSIM

◆ m_ScaleWeights

◆ m_SSIMPerScale

◆ m_StructureExponent

◆ MinimumScaleValue