Classes
struct	PCL_ROI

Enumerations
enum class	OrganizationType { Organized , Unorganized }

Functions
bool	vertSplit (const yarp::sig::Image &inImg, yarp::sig::Image &outImgL, yarp::sig::Image &outImgR)
	Split vertically an image in two images of the same size.

bool	horzSplit (const yarp::sig::Image &inImg, yarp::sig::Image &outImgUp, yarp::sig::Image &outImgDown)
	Split horizontally an image in two images of the same size.

bool	horzConcat (const yarp::sig::Image &inImgL, const yarp::sig::Image &inImgR, yarp::sig::Image &outImg)
	Concatenate horizontally two images of the same size in one with double width.

bool	vertConcat (const yarp::sig::Image &inImgUp, const yarp::sig::Image &inImgDown, yarp::sig::Image &outImg)
	Concatenate vertically two images of the same size in one with double height.

bool	cropRect (const yarp::sig::Image &inImg, const std::pair< unsigned int, unsigned int > &vertex1, const std::pair< unsigned int, unsigned int > &vertex2, yarp::sig::Image &outImg)
	Crop a rectangle area out of an image given two opposite vertices.

bool	sum (yarp::sig::Image &OutImg, const yarp::sig::Image &InImg, bool colorkey_enable, int colorkey, bool alpha_enable, float alpha, size_t off_x, size_t off_y)
	applies an image on the top over another image.

yarp::sig::PointCloud< yarp::sig::DataXYZ >	depthToPC (const yarp::sig::ImageOf< yarp::sig::PixelFloat > &depth, const yarp::sig::IntrinsicParams &intrinsic)
	depthToPC, compute the PointCloud given depth image and the intrinsic parameters of the camera.

yarp::sig::PointCloud< yarp::sig::DataXYZ >	depthToPC (const yarp::sig::ImageOf< yarp::sig::PixelFloat > &depth, const yarp::sig::IntrinsicParams &intrinsic, const yarp::sig::utils::PCL_ROI &roi, size_t step_x=1, size_t step_y=1, const std::string &output_order="+X+Y+Z")
	depthToPC, compute the PointCloud given depth image, the intrinsic parameters of the camera and a Region Of Interest.

template<typename T1 , typename T2 >
yarp::sig::PointCloud< T1 >	depthRgbToPC (const yarp::sig::ImageOf< yarp::sig::PixelFloat > &depth, const yarp::sig::ImageOf< T2 > &color, const yarp::sig::IntrinsicParams &intrinsic, const yarp::sig::utils::OrganizationType organizationType=yarp::sig::utils::OrganizationType::Organized, size_t step_x=1, size_t step_y=1, const std::string &output_order="+X+Y+Z")
	depthRgbToPC, compute the colored PointCloud given depth image, color image and the intrinsic parameters of the camera.

Enumeration Type Documentation

◆ OrganizationType

enum class yarp::sig::utils::OrganizationType

strong

Enumerator
Organized
Unorganized

Definition at line 15 of file PointCloudUtils.h.

Function Documentation

◆ cropRect()

bool yarp::sig::utils::cropRect	(	const yarp::sig::Image &	inImg,
		const std::pair< unsigned int, unsigned int > &	vertex1,
		const std::pair< unsigned int, unsigned int > &	vertex2,
		yarp::sig::Image &	outImg
	)

Crop a rectangle area out of an image given two opposite vertices.

Parameters

[in]	inImg	input image.
[in]	vertex1	first vertex of the crop rectangle area.
[in]	vertex2	second vertex of the crop rectangle area.
[out]	outImg	result of cropping the input image.

Note: Input and output images must have same pixel type and the crop area must lay within the input image. Vertices needn't be passed in the usual top-left, bottom-right order. The output image is resized to match the crop dimensions.

Returns: true on success, false otherwise.

Definition at line 115 of file ImageUtils.cpp.

◆ depthRgbToPC()

template<typename T1 , typename T2 >

yarp::sig::PointCloud< T1 > yarp::sig::utils::depthRgbToPC	(	const yarp::sig::ImageOf< yarp::sig::PixelFloat > &	depth,
		const yarp::sig::ImageOf< T2 > &	color,
		const yarp::sig::IntrinsicParams &	intrinsic,
		const yarp::sig::utils::OrganizationType	organizationType = `yarp::sig::utils::OrganizationType::Organized`,
		size_t	step_x = `1`,
		size_t	step_y = `1`,
		const std::string &	output_order = `"+X+Y+Z"`
	)

depthRgbToPC, compute the colored PointCloud given depth image, color image and the intrinsic parameters of the camera.

Parameters

[in]	depth,the	input depth image.
[in]	color,the	input color image.
[in]	intrinsic,intrinsic	parameter of the camera.
[in]	organizationType,if	organized the point cloud has size (width, depth). if unorganized the point cloud has size (width*height, 1). Note that in this case the data are organized column-wise (i.e. y,x and not x,y)
[in]	step_x,the	depth image size can be decimated, by selecting a column every step_x;
[in]	step_y,the	depth image size can be decimated, by selecting a row every step_y;
[in]	output_order,the	string will optionally rearrange data in any combination of positive and negative axes. e.g. "+Z-Y+X". Useful when dealing with not right-handed coordinate systems. Optimized for efficiency during point cloud creation with no additional computation.

Note: the intrinsic parameters are the one of the depth sensor if the depth frame IS NOT aligned with the colored one. On the other hand use the intrinsic parameters of the RGB camera if the frames are aligned.

Returns: the point cloud obtained by the de-projection.

Definition at line 62 of file PointCloudUtils-inl.h.

◆ depthToPC() [1/2]

PointCloud< DataXYZ > yarp::sig::utils::depthToPC	(	const yarp::sig::ImageOf< yarp::sig::PixelFloat > &	depth,
		const yarp::sig::IntrinsicParams &	intrinsic
	)

depthToPC, compute the PointCloud given depth image and the intrinsic parameters of the camera.

Parameters

[in]	depth,the	input depth image.
[in]	intrinsic,intrinsic	parameter of the camera.

Note: the intrinsic parameters are the one of the depth sensor if the depth frame IS NOT aligned with the colored one. On the other hand use the intrinsic parameters of the RGB camera if the frames are aligned.

Returns: the point cloud obtained by the de-projection.

Definition at line 16 of file PointCloudUtils.cpp.

◆ depthToPC() [2/2]

PointCloud< DataXYZ > yarp::sig::utils::depthToPC	(	const yarp::sig::ImageOf< yarp::sig::PixelFloat > &	depth,
		const yarp::sig::IntrinsicParams &	intrinsic,
		const yarp::sig::utils::PCL_ROI &	roi,
		size_t	step_x = `1`,
		size_t	step_y = `1`,
		const std::string &	output_order = `"+X+Y+Z"`
	)

depthToPC, compute the PointCloud given depth image, the intrinsic parameters of the camera and a Region Of Interest.

Parameters

[in]	depth,the	input depth image.
[in]	intrinsic,intrinsic	parameter of the camera.
[in]	roi,the	Region Of Interest intrinsic of the depth image that we want to convert.
[in]	step_x,the	depth image size can be decimated, by selecting a column every step_x;
[in]	step_y,the	depth image size can be decimated, by selecting a row every step_y;
[in]	output_order,the	string will optionally rearrange data in any combination of positive and negative axes. e.g. "+Z-Y+X". Useful when dealing with not right-handed coordinate systems. Optimized for efficiency during point cloud creation with no additional computation.

Note: the intrinsic parameters are the one of the depth sensor if the depth frame IS NOT aligned with the colored one. On the other hand use the intrinsic parameters of the RGB camera if the frames are aligned.

Returns: the point cloud obtained by the de-projection.

Definition at line 40 of file PointCloudUtils.cpp.

◆ horzConcat()

bool yarp::sig::utils::horzConcat	(	const yarp::sig::Image &	inImgL,
		const yarp::sig::Image &	inImgR,
		yarp::sig::Image &	outImg
	)

Concatenate horizontally two images of the same size in one with double width.

Parameters

[in]	inImgL	input left image.
[in]	inImgR	input right image.
[out]	outImg	result of the horizontal concatenation.

Note: The input images must have same dimensions and pixel type, and the output image must have same height and double width.

Returns: true on success, false otherwise.

Definition at line 69 of file ImageUtils.cpp.

◆ horzSplit()

bool yarp::sig::utils::horzSplit	(	const yarp::sig::Image &	inImg,
		yarp::sig::Image &	outImgUp,
		yarp::sig::Image &	outImgDown
	)

Split horizontally an image in two images of the same size.

Parameters

[in]	inImg	image to be horizontally split.
[out]	outImgUp	top half of inImg.
[out]	outImgDown	bottom half of inImg.

Note: The input image must have same height, double width of the output images and same pixel type.

Returns: true on success, false otherwise.

Definition at line 52 of file ImageUtils.cpp.

◆ sum()

bool yarp::sig::utils::sum	(	yarp::sig::Image &	OutImg,
		const yarp::sig::Image &	InImg,
		bool	colorkey_enable,
		int	colorkey,
		bool	alpha_enable,
		float	alpha,
		size_t	off_x,
		size_t	off_y
	)

applies an image on the top over another image.

Currently it is implemented only for RGB Images

Parameters

	[in/out]	OutImg the output image. It must be a valid image on the top of which data will be summed. It may contain a background or it can be zero.
[in]	InImg	the layer to be applied
[in]	colorkey	colorkey for the InImg image. If a pixel is == colorkey, then it will be made transparent and the background will be visible.
[in]	alpha	to be applied to InImg.
[in]	off_x	horizontal offset applied to InImg. Excess will be cropped.
[in]	off_y	vertical offset applied to InImg. Excess will be cropped.

Note: The two images must have the same pixelCode.

Returns: true on success, false if image cannot be summed because of incompatible format.

Definition at line 147 of file ImageUtils.cpp.

◆ vertConcat()

bool yarp::sig::utils::vertConcat	(	const yarp::sig::Image &	inImgUp,
		const yarp::sig::Image &	inImgDown,
		yarp::sig::Image &	outImg
	)

Concatenate vertically two images of the same size in one with double height.

Parameters

[in]	inImgUp	input top image.
[in]	inImgDown	input bottom image.
[out]	outImg	result of the horizontal concatenation.

Note: The input images must have same dimensions and pixel type, and the output image must have same width and double height.

Returns: true on success, false otherwise.

Definition at line 100 of file ImageUtils.cpp.

◆ vertSplit()

bool yarp::sig::utils::vertSplit	(	const yarp::sig::Image &	inImg,
		yarp::sig::Image &	outImgL,
		yarp::sig::Image &	outImgR
	)

Split vertically an image in two images of the same size.

Parameters

[in]	inImg	image to be vertically split.
[out]	outImgL	left half of inImg.
[out]	outImgR	right half of inImg.

Note: The input image must have same height, double width of the output images and same pixel type.

Returns: true on success, false otherwise.

Definition at line 25 of file ImageUtils.cpp.

Classes

Enumerations

Functions

Enumeration Type Documentation

◆ OrganizationType

Function Documentation

◆ cropRect()

◆ depthRgbToPC()

◆ depthToPC() [1/2]

◆ depthToPC() [2/2]

◆ horzConcat()

◆ horzSplit()

◆ sum()

◆ vertConcat()

◆ vertSplit()