Pixelwise Arithmetic Operations

Pixelwise arithmetic operations on 8-bit images. More...

Functions
int	rapp_pixop_set_u8 (uint8_t *buf, int dim, int width, int height, unsigned value)
	Set all pixels to a constant value.
int	rapp_pixop_not_u8 (uint8_t *buf, int dim, int width, int height)
	Negate all pixels.
int	rapp_pixop_flip_u8 (uint8_t *buf, int dim, int width, int height)
	Flip the sign bit.
int	rapp_pixop_lut_u8 (uint8_t *restrict buf, int dim, int width, int height, const uint8_t *restrict lut)
	Lookup-table transformation.
int	rapp_pixop_abs_u8 (uint8_t *buf, int dim, int width, int height)
	Absolute value.
int	rapp_pixop_addc_u8 (uint8_t *buf, int dim, int width, int height, int value)
	Add signed constant.
int	rapp_pixop_lerpc_u8 (uint8_t *buf, int dim, int width, int height, unsigned value, unsigned alpha8)
	Linear interpolation with constant.
int	rapp_pixop_lerpnc_u8 (uint8_t *buf, int dim, int width, int height, unsigned value, unsigned alpha8)
	Linear interpolation with a constant, rounded towards the constant value.
int	rapp_pixop_copy_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height)
	Copy all pixels.
int	rapp_pixop_add_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height)
	Saturated addition.
int	rapp_pixop_avg_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height)
	Average value.
int	rapp_pixop_sub_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height)
	Saturated subtraction.
int	rapp_pixop_subh_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height)
	Halved subtraction.
int	rapp_pixop_suba_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height)
	Absolute-value subtraction.
int	rapp_pixop_lerp_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height, unsigned alpha8)
	Linear interpolation.
int	rapp_pixop_lerpn_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height, unsigned alpha8)
	Linear interpolation with non-zero update, i.e., rounded towards src.
int	rapp_pixop_lerpi_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height, unsigned alpha8)
	Linear interpolation with inverted second operand, or generalized subtraction.
int	rapp_pixop_norm_u8 (uint8_t *restrict dst, int dst_dim, const uint8_t *restrict src, int src_dim, int width, int height)
	L1 norm.

---

Detailed Description

Pixelwise arithmetic operations on 8-bit images.

These functions operate in-place, i.e. the first operand is also the destination. All images must be aligned.

Next section: Thresholding

---

Function Documentation

int rapp_pixop_set_u8	(	uint8_t *	buf,
		int	dim,
		int	width,
		int	height,
		unsigned	value
	)

Set all pixels to a constant value.

Computes buf[i] = value.

Parameters:

[out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	value	The value to set, in the range 0 – 0xff.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_not_u8	(	uint8_t *	buf,
		int	dim,
		int	width,
		int	height
	)

Negate all pixels.

Computes buf[i] = 0xff - buf[i];

Parameters:

[in,out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_flip_u8	(	uint8_t *	buf,
		int	dim,
		int	width,
		int	height
	)

Flip the sign bit.

It is equivalant to converting between the formats two's complement and unsigned with bias. Computes buf[i] = buf[i] ^ 0x80;

Parameters:

[in,out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_lut_u8	(	uint8_t *restrict	buf,
		int	dim,
		int	width,
		int	height,
		const uint8_t *restrict	lut
	)

Lookup-table transformation.

Computes buf[i] = lut[buf[i]].

Parameters:

[in,out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	lut	The 8-bit lookup table to use, with at least 256 bytes. It may be misaligned.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_abs_u8	(	uint8_t *	buf,
		int	dim,
		int	width,
		int	height
	)

Absolute value.

Computes 2*abs(buf[i] - 0x80). The result is saturated.

Parameters:

[in,out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_addc_u8	(	uint8_t *	buf,
		int	dim,
		int	width,
		int	height,
		int	value
	)

Add signed constant.

Computes buf[i] = buf[i] + value. The result is saturated.

Parameters:

[in,out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	value	The value to add.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_lerpc_u8	(	uint8_t *	buf,
		int	dim,
		int	width,
		int	height,
		unsigned	value,
		unsigned	alpha8
	)

Linear interpolation with constant.

Computes buf[i] += alpha8*(value - buf[i]), where the multiplication is a rounded 8-bit fixed-point multiply.

Parameters:

[in,out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	value	The value to interpolate with, in the range 0 – 0xff.
	alpha8	The Q.8 fixed-point blend factor, in the range 0 – 0x100.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_lerpnc_u8	(	uint8_t *	buf,
		int	dim,
		int	width,
		int	height,
		unsigned	value,
		unsigned	alpha8
	)

Linear interpolation with a constant, rounded towards the constant value.

Computes buf[i] += alpha8*(value - buf[i]), where the multiplication is an 8-bit fixed-point multiply, rounded away from zero.

Parameters:

[in,out]	buf	Pixel buffer pointer.
	dim	Pixel buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	value	The value to interpolate with, in the range 0 – 0xff.
	alpha8	The Q.8 fixed-point blend factor, in the range 0 – 0x100.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_copy_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height
	)

Copy all pixels.

Computes dst[i] = src[i].

Parameters:

[out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_add_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height
	)

Saturated addition.

Computes dst[i] += src[i]. The result is saturated.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_avg_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height
	)

Average value.

Computes dst[i] = (dst[i] + src[i] + 1) / 2.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_sub_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height
	)

Saturated subtraction.

Computes dst[i] -= src[i]. The result is saturated.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_subh_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height
	)

Halved subtraction.

Computes dst[i] = (dst[i] - src[i] + 0x100) / 2.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_suba_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height
	)

Absolute-value subtraction.

Computes dst[i] = abs(dst[i] - src[i]).

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_lerp_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height,
		unsigned	alpha8
	)

Linear interpolation.

Computes dst[i] += alpha8*(src[i] - dst[i]), where the multiplication is a rounded 8-bit fixed-point multiply.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	alpha8	The Q.8 fixed-point blend factor, in the range 0 – 0x100.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_lerpn_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height,
		unsigned	alpha8
	)

Linear interpolation with non-zero update, i.e., rounded towards src.

Computes dst[i] += alpha8*(src[i] - dst[i]), where the multiplication is an 8-bit fixed-point multiply, rounded away from zero.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	alpha8	The Q.8 fixed-point blend factor, in the range 0 – 0x100.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_lerpi_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height,
		unsigned	alpha8
	)

Linear interpolation with inverted second operand, or generalized subtraction.

Computes dst[i] += alpha8*(0xff - src[i] - dst[i]), where the multiplication is a rounded 8-bit fixed-point multiply.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.
	alpha8	The Q.8 fixed-point blend factor, in the range 0 – 0x100.

Returns:

A negative error code on error, zero otherwise.

int rapp_pixop_norm_u8	(	uint8_t *restrict	dst,
		int	dst_dim,
		const uint8_t *restrict	src,
		int	src_dim,
		int	width,
		int	height
	)

L1 norm.

Computes dst[i] = (abs(dst[i]) + abs(src[i]) + 1) / 2.

Parameters:

[in,out]	dst	Destination buffer pointer.
	dst_dim	Destination buffer row dimension in bytes.
[in]	src	Source buffer pointer.
	src_dim	Source buffer row dimension in bytes.
	width	Image width in pixels.
	height	Image height in pixels.

Returns:

A negative error code on error, zero otherwise.