TLIB documentation
 

tlConvolution.h File Reference

---

Detailed Description

Internal convolution operators.

#include "tlVision.h"

Include dependency graph for tlConvolution.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Functions
int	tl_convolution (tlPixel *src, int width, int height, int pixel_width, char *kernel, int kernel_width, int kernel_height, tlRect *rect, tlPixel *dest)
int	tl_convolution (tlPixel *src, int width, int height, int pixel_width, float *kernel, int kernel_width, int kernel_height, tlRect *rect, tlPixel *dest)
int	tl_convolution_x (tlPixel *src, int width, int height, int pixel_width, char *kernel, int kernel_size, tlRect *rect, int *dest)
int	tl_convolution_x (tlPixel *src, int width, int height, int pixel_width, float *kernel, int kernel_size, tlRect *rect, float *dest)
int	tl_convolution_x (tlPixel *src, int width, int height, int pixel_width, float *kernel, int kernel_size, tlRect *rect, tlPixel *mask, float *dest)
int	tl_convolution_y (int *src, int width, int height, int pixel_width, char *kernel, int kernel_size, tlRect *rect, tlPixel *dest)
int	tl_convolution_y (float *src, int width, int height, int pixel_width, float *kernel, int kernel_size, tlRect *rect, tlPixel *dest)
int	tl_convolution_y (float *src, int width, int height, int pixel_width, float *kernel, int kernel_size, tlRect *rect, tlPixel *mask, tlPixel *dest)
int	tl_create_gaussian_kernel (int ksize, float sigma, float *kernel)

---

Function Documentation

int tl_convolution	(	tlPixel *	src,
		int	width,
		int	height,
		int	pixel_width,
		float *	kernel,
		int	kernel_width,
		int	kernel_height,
		tlRect *	rect,
		tlPixel *	dest
	)

Perform linear convolution with any square kernel (with floating-point values).

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (square)
	kernel_width	width of the convolution kernel
	kernel_height	height of the convolution kernel
	rect	portion of the image to process
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_convolution	(	tlPixel *	src,
		int	width,
		int	height,
		int	pixel_width,
		char *	kernel,
		int	kernel_width,
		int	kernel_height,
		tlRect *	rect,
		tlPixel *	dest
	)

Perform linear convolution with any square kernel (with integer values).

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (square)
	kernel_width	width of the convolution kernel
	kernel_height	height of the convolution kernel
	rect	portion of the image to process
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_convolution_x	(	tlPixel *	src,
		int	width,
		int	height,
		int	pixel_width,
		float *	kernel,
		int	kernel_size,
		tlRect *	rect,
		tlPixel *	mask,
		float *	dest
	)

Perform linear convolution with any square kernel (with floating-point values); this routine performs monodimensional convolution along all rows of an image with a separable kernel

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (line)
	kernel_size	size of the convolution kernel
	rect	portion of the image to process
	mask	Mask the object of the src (must be of the same size of the source)
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_convolution_x	(	tlPixel *	src,
		int	width,
		int	height,
		int	pixel_width,
		float *	kernel,
		int	kernel_size,
		tlRect *	rect,
		float *	dest
	)

Perform linear convolution with any square kernel (with floating-point values); this routine performs monodimensional convolution along all lines of an image with a separable kernel

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (line)
	kernel_size	size of the convolution kernel
	rect	portion of the image to process
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_convolution_x	(	tlPixel *	src,
		int	width,
		int	height,
		int	pixel_width,
		char *	kernel,
		int	kernel_size,
		tlRect *	rect,
		int *	dest
	)

Perform linear convolution with any square kernel (with integer values); this routine performs monodimensional convolution along all lines of an image with a separable kernel

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (line)
	kernel_size	size of the convolution kernel
	rect	portion of the image to process
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_convolution_y	(	float *	src,
		int	width,
		int	height,
		int	pixel_width,
		float *	kernel,
		int	kernel_size,
		tlRect *	rect,
		tlPixel *	mask,
		tlPixel *	dest
	)

Perform linear convolution with any square kernel (with floating-point values); this routine performs monodimensional convolution along all columns of an image with a separable kernel

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (line)
	kernel_size	size of the convolution kernel
	rect	portion of the image to process
	mask	mask the object of the src (must be of the same size of the source)
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_convolution_y	(	float *	src,
		int	width,
		int	height,
		int	pixel_width,
		float *	kernel,
		int	kernel_size,
		tlRect *	rect,
		tlPixel *	dest
	)

Perform linear convolution with any square kernel (with floating-point values); this routine performs monodimensional convolution along all columns of an image with a separable kernel

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (line)
	kernel_size	size of the convolution kernel
	rect	portion of the image to process
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_convolution_y	(	int *	src,
		int	width,
		int	height,
		int	pixel_width,
		char *	kernel,
		int	kernel_size,
		tlRect *	rect,
		tlPixel *	dest
	)

Perform linear convolution with any square kernel (with integer values); this routine performs monodimensional convolution along all columns of an image with a separable kernel

Parameters:

	src	image pixel array
	width	image width
	height	image height
	pixel_width	pixel width
	kernel	convolution kernel (line)
	kernel_size	size of the convolution kernel
	rect	portion of the image to process
	dest	[out] destination pixel array

Returns:

0 on success, -1 otherwise.
See error management for details.

int tl_create_gaussian_kernel	(	int	ksize,
		float	sigma,
		float *	kernel
	)

Compute gaussian kernel of a given size. The kernel is a line and must be applied by using the convolution in separate mode.

See also:

tlPixelArray::convolution(float *xkernel, int xksize, float *ykernel, int yksize)

Parameters:

	ksize	kernel size (line)
	sigma	desired sigma
	kernel	[out] destination gaussian kernel array

Returns:

0 on success, -1 otherwise.
See error management for details.

Todo:

check why output is different from Gaetan's version

---

TLIB documentation - generated on 7 May 2008
Please address any questions to seb@tuyphon.com
(C) 2001-2008 - S. Grange
(C) 2001-2007 - VRAI Group, EPFL
All Rights Reserved.