Class with operations on image objects; Each of these operations creates a new image object that the caller becomes responsable for. More...

#include <ImageTools.h>

Public Member Functions
	ImageTools ()
	constructor

	~ImageTools ()
	destructor

Static Public Member Functions
static Image *	MirrorBorder (Image *pInImage, int borderWidth, int borderHeight)
	Creates from source image a new image with added mirrored borders The size of the border is borderWidth pixels left and right and borderHeight top and bottom. More...

static Image *	CropBorder (Image *pInImage, int borderWidth, int borderHeight)
	Crops the image by cutting a border from around it of borderWidth pixels left and right and borderHeight top and bottom. More...

static Image *	PaddBorder (Image *pInImage, int borderWidth, int borderHeight, double paddingValue)
	Creates from source image a new image with added padded borders The size of the border is borderWidth pixels left and right and borderHeight top and bottom. More...

static Image *	ExtractSubImage (Image *pInImage, common::Point< int > topLeft, common::Point< int > bottomRight)
	Creates new subimage cropped from current image. More...

static bool	InsertSubGrid (Image pImage, ArrayGrid< double > pSubGrid, int xTop, int yTop)
	overwrites part of an image with a subimage More...

static void	Diagnose (Image *pImage, std::string ID=std::string(""))
	prints some image properties on std out for diagnostics More...

static Image *	CreateImageSSD (Image pImage1, Image pImage2, bool printOutput=false)
	creates new difference image between two images (squared difference measure) More...

static double	ComputeMSE (Image pImage1, Image pImage2)
	computes MSE (Mean Square Error) between two images More...

static double	ComputePSNR (Image pImage1, Image pImage2)
	computes PSNR (Peak Signal to noise Ratio) between two images More...

static Image *	CreateImageSSIM (Image pImage1, Image pImage2, int localWindowSize=9)
	creates new difference image between two images (Structural SIMilarity measure) More...

static std::vector< double >	GetColorMappingFactors (Image pImage1, Image pImage2)
	help function to adjust global color perception of one image compared to another More...

static Image *	CreateCheckeredImage (Image pImage1, Image pImage2, int blockSize=35)
	creates a new checkerboard image of the two input images Displays squares of size blockSize of alternating the first and the second image More...

static Image *	CreateTransparantlyMixedImage (Image pImage1, Image pImage2, double thisWeight=0.5)
	creates a new transparant image from a weighted average of the two input images More...

static Image *	ApplyGamma (Image *pInImage, double gamma)
	creates new image with adjusted gamma More...

static Image *	CreateLinearRescaledImage (Image *pImage, bool bandsIndependent)
	creates a new linear rescaled image of an input image More...

static Image *	ApplyJetColorMap (Image *pInImage)
	Creates a new image in false color using the "jet" color mapping This false color mapping for gray value images is well known from usage in Matlab. It ranges from blue for dark intensities to red for bright intensities, and passes through the colors cyan, yellow, and orange. More...

static Image *	ConvertToGrayImage (Image *pInImage)
	Converts current image to gray image. More...

static Image *	ConvertToSepiaImage (Image *pInImage)
	Converts current image to sepia-toned image. More...

static Image *	Negative (Image *pInImage)
	Converts current image to its negative image Cfr photographic negative; each values is replaced by the maximum allowed value minus the current value. More...

static Image *	Rotate90DegreesClockwise (Image *pInImage)
	Creates a new image: rotates input image 90 degrees clockwise. More...

static Image *	Rotate90DegreesCounterClockwise (Image *pInImage)
	Creates a new image: rotates input image 90 degrees counterclockwise. More...

static Image *	Rotate180Degrees (Image *pInImage)
	Creates a new image: rotates input image 180 degrees. More...

static double	GetLocalDarkChannel (Image *pImage, int xLocalCenter, int yLocalCenter, int windowSize)
	determines the minimum value in a local window over all spectral bands More...

static unsigned int *	CreateIntArrayFromColorImage (Image *pImage)
	creates an array of 32bit unsigned int values, where each int value combines the three color band values per pixel (each 8bit) in one 32bit value [ 0 0 0 0 0 0 0 0 0 ] [ 1 1 1 1 1 1 1 1 ] [ 1 1 1 1 1 1 1 1 ] [ 1 1 1 1 1 1 1 1 ] unused value R value G value B More...

static Image *	CreateColorImageFromIntArray (unsigned int *pIntArray, int width, int height)
	creates an image from an array of 32bit integers by splitting again the integers in unsigned chars per band value More...

Detailed Description

Class with operations on image objects; Each of these operations creates a new image object that the caller becomes responsable for.

The methods in this class are grouped thematically in the following categories

CONSTRUCTOR / DESTRUCTOR
PADD / CROP IMAGE
COMPARE IMAGES
MIX TWO IMAGES
REMAP IMAGE INTENSITIES
ROTATE IMAGE LOSSLESS (ANGLE IS MULTIPLE OF 90 DEGREES)
MISCELLANEOUS

Member Function Documentation

Image * stira::imagetools::ImageTools::ApplyGamma	(	Image *	pInImage,
		double	gamma
	)

static

creates new image with adjusted gamma

Parameters

pInImage	input image
gamma	gamma factor to apply

References stira::imagedata::Image::Clone(), stira::common::MathUtils::GammaCorrect(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetImageName(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::Image::GetWidth(), and stira::imagedata::Image::SetImageName().

Image * stira::imagetools::ImageTools::ApplyJetColorMap ( Image * pInImage )

static

Creates a new image in false color using the "jet" color mapping This false color mapping for gray value images is well known from usage in Matlab. It ranges from blue for dark intensities to red for bright intensities, and passes through the colors cyan, yellow, and orange.

Returns: RGB image according to jet color map

References stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::common::MathUtils::GetMax(), stira::common::MathUtils::GetMin(), stira::imagedata::GridStatistics< T >::GetMinMax(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::ArrayGrid< T >::GetValue(), stira::imagedata::Image::GetWidth(), and stira::imagedata::Image::SetImageName().

double stira::imagetools::ImageTools::ComputeMSE	(	Image *	pImage1,
		Image *	pImage2
	)

static

computes MSE (Mean Square Error) between two images

Parameters

pImage1	first image
pImage2	second image

Returns: MSE value

References stira::imagetools::NumberGridTools< T >::ComputeMSE(), stira::imagedata::Image::GetBands(), and stira::imagedata::Image::GetNumberOfBands().

double stira::imagetools::ImageTools::ComputePSNR	(	Image *	pImage1,
		Image *	pImage2
	)

static

computes PSNR (Peak Signal to noise Ratio) between two images

Parameters

pImage1	first image
pImage2	second image

Returns: PSNR value

Image * stira::imagetools::ImageTools::ConvertToGrayImage ( Image * pInImage )

static

Converts current image to gray image.

Parameters

pInImage input image This is just the simplest possible technique; to be extended later with e.g., projection on a PCA basis

References stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetColor(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::Image::GetWidth(), stira::imagedata::TransformColorSpace::RGBToGray(), and stira::imagedata::Image::SetImageName().

Referenced by ConvertToSepiaImage().

Image * stira::imagetools::ImageTools::ConvertToSepiaImage ( Image * pInImage )

static

Converts current image to sepia-toned image.

Parameters

pInImage input image

References ConvertToGrayImage(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::Image::GetWidth(), stira::imagedata::TransformColorSpace::RGBToSepia2(), stira::imagedata::Image::SetColor(), and stira::imagedata::Image::SetImageName().

Image * stira::imagetools::ImageTools::CreateCheckeredImage	(	Image *	pImage1,
		Image *	pImage2,
		int	blockSize = `35`
	)

static

creates a new checkerboard image of the two input images Displays squares of size blockSize of alternating the first and the second image

Parameters

pImage1	first image
pImage2	second image
blockSize	size of alternating blocks

Returns: checker board image; 0 if images have different dimensions or nr of bands

References stira::imagedata::Image::Clone(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetImageName(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::ArrayGrid< T >::GetValue(), stira::imagedata::Image::GetWidth(), stira::imagedata::Image::SetImageName(), and stira::imagedata::ArrayGrid< T >::SetValue().

Image * stira::imagetools::ImageTools::CreateColorImageFromIntArray	(	unsigned int *	pIntArray,
		int	width,
		int	height
	)

static

creates an image from an array of 32bit integers by splitting again the integers in unsigned chars per band value

Parameters

pIntArray	the input array of 32bit integers
width	the width of the target image
height	the height of the target image

References stira::imagedata::Image::SetColor(), and stira::common::MathUtils::SplitIntInChars().

Image * stira::imagetools::ImageTools::CreateImageSSD	(	Image *	pImage1,
		Image *	pImage2,
		bool	printOutput = `false`
	)

static

creates new difference image between two images (squared difference measure)

Parameters

pImage1	first image
pImage2	second image
printOutput	flag if output needs to be printed to console

Returns: image with visualization of squared difference

References stira::imagedata::Image::AddBand(), stira::imagetools::NumberGridTools< T >::CreateSquaredErrorGrid(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetImageName(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::Image::GetWidth(), and stira::imagedata::Image::SetImageName().

Image * stira::imagetools::ImageTools::CreateImageSSIM	(	Image *	pImage1,
		Image *	pImage2,
		int	localWindowSize = `9`
	)

static

creates new difference image between two images (Structural SIMilarity measure)

Parameters

pImage1	first image
pImage2	second image

Returns: image with visualization of Structural SIMilarity measure

References stira::imagedata::Image::AddBand(), stira::imagetools::NumberGridTools< T >::ComputeSSIM(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetImageName(), stira::imagedata::GridStatistics< T >::GetMinMax(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::Image::GetWidth(), and stira::imagedata::Image::SetImageName().

unsigned int * stira::imagetools::ImageTools::CreateIntArrayFromColorImage ( Image * pImage )

static

creates an array of 32bit unsigned int values, where each int value combines the three color band values per pixel (each 8bit) in one 32bit value [ 0 0 0 0 0 0 0 0 0 ] [ 1 1 1 1 1 1 1 1 ] [ 1 1 1 1 1 1 1 1 ] [ 1 1 1 1 1 1 1 1 ] unused value R value G value B

Parameters

pImage input image, assumes that values per band internally can be cast to unsigned char

References stira::imagedata::ColorValue::c, stira::common::MathUtils::CombineCharsInInt(), stira::imagedata::Image::GetColor(), stira::imagedata::Image::GetHeight(), and stira::imagedata::Image::GetWidth().

Image * stira::imagetools::ImageTools::CreateLinearRescaledImage	(	Image *	pImage,
		bool	bandsIndependent
	)

static

creates a new linear rescaled image of an input image

Parameters

pImage1	first image
pImage2	second image
bandsIndependent	flag if all bands are rescaled independant of each other

Returns: linearly rescaled image

References stira::imagedata::Image::Clone(), stira::imagedata::Image::GetBands(), stira::imagedata::GridStatistics< T >::GetMinMax(), and stira::imagedata::Image::GetNumberOfBands().

Image * stira::imagetools::ImageTools::CreateTransparantlyMixedImage	(	Image *	pImage1,
		Image *	pImage2,
		double	thisWeight = `0.5`
	)

static

creates a new transparant image from a weighted average of the two input images

Parameters

pImage1	first image
pImage2	second image
thisWeight	weight for the current image (must be > 0 and < 1; else 0.5 is used)

Returns: transparant image; 0 if images have different dimensions or nr of bands

References stira::imagedata::Image::Clone(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetImageName(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::ArrayGrid< T >::GetValue(), stira::imagedata::Image::GetWidth(), stira::imagedata::Image::SetImageName(), and stira::imagedata::ArrayGrid< T >::SetValue().

Image * stira::imagetools::ImageTools::CropBorder	(	Image *	pInImage,
		int	borderWidth,
		int	borderHeight
	)

static

Crops the image by cutting a border from around it of borderWidth pixels left and right and borderHeight top and bottom.

Parameters

pInImage	input image
borderWidth	width of the extra border left and right
borderHeight	height of the extra border top and bottom

Returns: new image (responsability of caller to delete) or 0 when mirroring couldn't be executed

References stira::imagedata::Image::AddBand(), stira::imagedata::GridExtender< T >::CropBorder(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), and stira::imagedata::Image::GetWidth().

Referenced by stira::filter::GaussConvolve::DerivativeConvolveFFT(), stira::fouriertools::FFT::GaussConvolve(), stira::filter::KuwaharaFilter::Run(), stira::filter::LaplacianOfGaussianFilter::Run(), stira::filter::AdaptiveBilateralFilter::Run(), stira::filter::NonLocalMeansFilter::Run(), stira::filter::LocalBinaryPattern::RunClassic(), stira::filter::MedianFilter::RunHybridMedian(), and stira::filter::MedianFilter::RunMedian().

void stira::imagetools::ImageTools::Diagnose	(	Image *	pImage,
		std::string	ID = `std::string("")`
	)

static

prints some image properties on std out for diagnostics

Parameters

ID	identifier which image is being diagnosed

References stira::imagedata::Image::GetBands(), stira::imagedata::GridStatistics< T >::GetGridKurtosis(), stira::imagedata::GridStatistics< T >::GetGridMean(), stira::imagedata::GridStatistics< T >::GetGridVariance(), stira::imagedata::Image::GetHeight(), stira::imagedata::GridStatistics< T >::GetMinMax(), stira::imagedata::Image::GetNumberOfBands(), and stira::imagedata::Image::GetWidth().

Image * stira::imagetools::ImageTools::ExtractSubImage	(	Image *	pInImage,
		common::Point< int >	topLeft,
		common::Point< int >	bottomRight
	)

static

Creates new subimage cropped from current image.

Parameters

pInImage	input image
topLeft	top left corner of region to crop
bottomRight	bottom right corner of region to crop

Returns: new image (responsability of caller to delete) or 0 when mirroring couldn't be executed

References stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::Image::GetWidth(), stira::common::Point< T >::x, and stira::common::Point< T >::y.

std::vector< double > stira::imagetools::ImageTools::GetColorMappingFactors	(	Image *	pImage1,
		Image *	pImage2
	)

static

help function to adjust global color perception of one image compared to another

Parameters

pImage1	reference image
pImage2	image to compare with reference image

Returns: vector of ratios of color channel means

References stira::imagedata::Image::GetBands(), stira::imagedata::GridStatistics< T >::GetGridMean(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), and stira::imagedata::Image::GetWidth().

double stira::imagetools::ImageTools::GetLocalDarkChannel	(	Image *	pImage,
		int	xLocalCenter,
		int	yLocalCenter,
		int	windowSize
	)

static

determines the minimum value in a local window over all spectral bands

Parameters

pImage	image to investigate
xLocalCenter	x coordinate of local patch under consideration
yLocalCenter	y coordinate of local patch under consideration
windowSize	size of local patches to take

References stira::imagedata::Image::GetBands(), stira::imagedata::GridStatistics< T >::GetLocalMinimum(), and stira::imagedata::Image::GetNumberOfBands().

Referenced by stira::contrastenhance::HazeRemover::Run(), and stira::contrastenhance::HazeRemover::~HazeRemover().

bool stira::imagetools::ImageTools::InsertSubGrid	(	Image *	pImage,
		ArrayGrid< double > *	pSubGrid,
		int	xTop,
		int	yTop
	)

static

overwrites part of an image with a subimage

Parameters

pSubGrid	grid to insert
xTop	x coordinate in parent image where to insert top left corner of sub image
yTop	y coordinate in parent image where to insert top left corner of sub image

References stira::imagedata::ArrayGrid< T >::Clone(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::ArrayGrid< T >::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::ArrayGrid< T >::GetValue(), stira::imagedata::Image::GetWidth(), and stira::imagedata::ArrayGrid< T >::GetWidth().

Referenced by stira::imagetools::PyramidTools::VisualizeRealPyramid().

Image * stira::imagetools::ImageTools::MirrorBorder	(	Image *	pInImage,
		int	borderWidth,
		int	borderHeight
	)

static

Creates from source image a new image with added mirrored borders The size of the border is borderWidth pixels left and right and borderHeight top and bottom.

Parameters

pInImage	input image
borderWidth	width of the extra border left and right
borderHeight	height of the extra border top and bottom

Returns: new image (responsability of caller to delete) or 0 when mirroring couldn't be executed

References stira::imagedata::Image::AddBand(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::Image::GetWidth(), and stira::imagedata::GridExtender< T >::MirrorBorder().

Referenced by stira::filter::GaussConvolve::DerivativeConvolveFFT(), stira::fouriertools::FFT::GaussConvolve(), stira::filter::KuwaharaFilter::Run(), stira::filter::LaplacianOfGaussianFilter::Run(), stira::filter::AdaptiveBilateralFilter::Run(), stira::filter::NonLocalMeansFilter::Run(), stira::filter::LocalBinaryPattern::RunClassic(), stira::filter::MedianFilter::RunHybridMedian(), and stira::filter::MedianFilter::RunMedian().

Image * stira::imagetools::ImageTools::Negative ( Image * pInImage )

static

Converts current image to its negative image Cfr photographic negative; each values is replaced by the maximum allowed value minus the current value.

Parameters

pInImage input image

References stira::imagedata::Image::Clone(), stira::imagedata::Image::GetBands(), stira::imagedata::Image::GetHeight(), stira::imagedata::Image::GetNumberOfBands(), stira::imagedata::ArrayGrid< T >::GetValue(), stira::imagedata::Image::GetWidth(), and stira::imagedata::ArrayGrid< T >::SetValue().

Image * stira::imagetools::ImageTools::PaddBorder	(	Image *	pInImage,
		int	borderWidth,
		int	borderHeight,
		double	paddingValue
	)

static

Creates from source image a new image with added padded borders The size of the border is borderWidth pixels left and right and borderHeight top and bottom.

Parameters

pInImage	input image
borderWidth	width of the extra border left and right
borderHeight	height of the extra border top and bottom
paddingValue	value to be used for padding