Operations

Various mathematical operations with images. More...

Functions
SIMD_API void	SimdOperationBinary8u (const uint8_t *a, size_t aStride, const uint8_t *b, size_t bStride, size_t width, size_t height, size_t channelCount, uint8_t *dst, size_t dstStride, SimdOperationBinary8uType type)
	Performs an element-wise binary operation between two 8-bit images. More...
SIMD_API void	SimdOperationBinary16i (const uint8_t *a, size_t aStride, const uint8_t *b, size_t bStride, size_t width, size_t height, uint8_t *dst, size_t dstStride, SimdOperationBinary16iType type)
	Performs an element-wise binary operation between two signed 16-bit images. More...
SIMD_API void	SimdVectorProduct (const uint8_t *vertical, const uint8_t *horizontal, uint8_t *dst, size_t stride, size_t width, size_t height)
	Calculates an 8-bit gray image as the normalized outer product of two 8-bit vectors. More...
template<template< class > class A>
SIMD_INLINE void	OperationBinary8u (const View< A > &a, const View< A > &b, View< A > &dst, SimdOperationBinary8uType type)
	Performs given operation between two images. More...
template<template< class > class A>
SIMD_INLINE void	OperationBinary16i (const View< A > &a, const View< A > &b, View< A > &dst, SimdOperationBinary16iType type)
	Performs given operation between two images. More...
template<template< class > class A>
SIMD_INLINE void	VectorProduct (const uint8_t *vertical, const uint8_t *horizontal, View< A > &dst)
	Calculates result 8-bit gray image as product of two vectors. More...

Detailed Description

Various mathematical operations with images.

Function Documentation

SimdOperationBinary8u()

void SimdOperationBinary8u	(	const uint8_t *	a,
		size_t	aStride,
		const uint8_t *	b,
		size_t	bStride,
		size_t	width,
		size_t	height,
		size_t	channelCount,
		uint8_t *	dst,
		size_t	dstStride,
		SimdOperationBinary8uType	type
	)

Performs an element-wise binary operation between two 8-bit images.

All images must have the same width, height and number of channels. The function processes width*channelCount unsigned 8-bit values in every row; every channel is handled independently. The exact operation is selected by type (average, bitwise AND/OR, maximum, minimum, saturated subtraction or saturated addition).

Note

This function has a C++ wrapper: Simd::OperationBinary8u(const View<A>& a, const View<A>& b, View<A>& dst, SimdOperationBinary8uType type).

Parameters

[in]	a	- a pointer to pixels data of the first input image.
[in]	aStride	- a row size of the first image.
[in]	b	- a pointer to pixels data of the second input image.
[in]	bStride	- a row size of the second image.
[in]	width	- an image width.
[in]	height	- an image height.
[in]	channelCount	- a number of 8-bit channels per pixel.
[out]	dst	- a pointer to pixels data of the output image.
[in]	dstStride	- a row size of the output image.
[in]	type	- a type of operation (see SimdOperationBinary8uType).

SimdOperationBinary16i()

void SimdOperationBinary16i	(	const uint8_t *	a,
		size_t	aStride,
		const uint8_t *	b,
		size_t	bStride,
		size_t	width,
		size_t	height,
		uint8_t *	dst,
		size_t	dstStride,
		SimdOperationBinary16iType	type
	)

Performs an element-wise binary operation between two signed 16-bit images.

All images must have the same width, height and SimdPixelFormatInt16 pixel format. The function treats every row as an array of width signed 16-bit values and applies the non-saturated operation selected by type (addition or subtraction). Strides are specified in bytes and must be multiples of sizeof(int16_t).

Note

This function has a C++ wrapper: Simd::OperationBinary16i(const View<A>& a, const View<A>& b, View<A>& dst, SimdOperationBinary16iType type).

Parameters

[in]	a	- a pointer to pixels data of the first input image.
[in]	aStride	- a row size of the first image.
[in]	b	- a pointer to pixels data of the second input image.
[in]	bStride	- a row size of the second image.
[in]	width	- an image width.
[in]	height	- an image height.
[out]	dst	- a pointer to pixels data of the output image.
[in]	dstStride	- a row size of the output image.
[in]	type	- a type of operation (see SimdOperationBinary16iType).

SimdVectorProduct()

void SimdVectorProduct	(	const uint8_t *	vertical,
		const uint8_t *	horizontal,
		uint8_t *	dst,
		size_t	stride,
		size_t	width,
		size_t	height
	)

Calculates an 8-bit gray image as the normalized outer product of two 8-bit vectors.

For all points:

dst[x, y] = DivideBy255(horizontal[x]*vertical[y]);
where DivideBy255(v) = (v + 1 + (v >> 8)) >> 8.

Note

This function has a C++ wrapper: Simd::VectorProduct(const uint8_t * vertical, const uint8_t * horizontal, View<A>& dst).

Parameters

[in]	vertical	- a pointer to the vertical vector. Its length must be equal to the output image height.
[in]	horizontal	- a pointer to the horizontal vector. Its length must be equal to the output image width.
[out]	dst	- a pointer to pixels data of the output image.
[in]	stride	- a row size of the output image in bytes.
[in]	width	- a width of the output image.
[in]	height	- a height of the output image.

OperationBinary8u()

void OperationBinary8u	(	const View< A > &	a,
		const View< A > &	b,
		View< A > &	dst,
		SimdOperationBinary8uType	type
	)

Performs given operation between two images.

All images must have the same width, height and format (8-bit gray, 16-bit UV (UV plane of NV12 pixel format), 24-bit BGR or 32-bit BGRA).

Note

This function is a C++ wrapper for function SimdOperationBinary8u.

Parameters

[in]	a	- a first input image.
[in]	b	- a second input image.
[out]	dst	- an output image.
[in]	type	- a type of operation (see SimdOperationBinary8uType).

OperationBinary16i()

void OperationBinary16i	(	const View< A > &	a,
		const View< A > &	b,
		View< A > &	dst,
		SimdOperationBinary16iType	type
	)

Performs given operation between two images.

All images must have the same width, height and Simd::View::Int16 pixel format.

Note

This function is a C++ wrapper for function SimdOperationBinary16i.

Parameters

[in]	a	- a first input image.
[in]	b	- a second input image.
[out]	dst	- an output image.
[in]	type	- a type of operation (see SimdOperationBinary16iType).

VectorProduct()

void VectorProduct	(	const uint8_t *	vertical,
		const uint8_t *	horizontal,
		View< A > &	dst
	)

Calculates result 8-bit gray image as product of two vectors.

For all points:

dst[x, y] = horizontal[x]*vertical[y]/255;

Note

This function is a C++ wrapper for function SimdVectorProduct.

Parameters

[in]	vertical	- a pointer to pixels data of vertical vector. It length is equal to result image height.
[in]	horizontal	- a pointer to pixels data of horizontal vector. It length is equal to result image width.
[out]	dst	- a result image.