VXLIB_gaussian_5x5_i8u_o8u

Functions
VXLIB_STATUS	VXLIB_gaussian_5x5_i8u_o8u (const uint8_t src[restrict], const VXLIB_bufParams2D_t *src_addr, uint8_t dst[restrict], const VXLIB_bufParams2D_t *dst_addr, uint8_t shift)
VXLIB_STATUS	VXLIB_gaussian_5x5_i8u_o8u_checkParams (const uint8_t src[restrict], const VXLIB_bufParams2D_t *src_addr, const uint8_t dst[restrict], const VXLIB_bufParams2D_t *dst_addr, uint8_t shift)

Detailed Description

Function Documentation

VXLIB_gaussian_5x5_i8u_o8u()

VXLIB_STATUS VXLIB_gaussian_5x5_i8u_o8u	(	const uint8_t	src[restrict],
		const VXLIB_bufParams2D_t *	src_addr,
		uint8_t	dst[restrict],
		const VXLIB_bufParams2D_t *	dst_addr,
		uint8_t	shift
	)

Description:

Computes a gaussian filter over the 5x5 neighborhood of each input pixel.

Method:

This filter uses the following convolution matrix:

         | 1  4  6  4  1 |
         | 4 16 24 16  4 |
    K =  | 6 24 36 24  6 | >> shift (shift is typically 8 to divide the result by 256)
         | 4 16 24 16  4 |
         | 1  4  6  4  1 |

Parameters

[in]	src[]	Pointer to array containing input image (UQ8.0)
[in]	src_addr[]	Pointer to structure containing dimensional information of src
[out]	dst[]	Pointer to array containing output image (UQ8.0)
[in]	dst_addr[]	Pointer to structure containing dimensional information of dst
[in]	shift	Right shift value after the sum of products of the 25 neighborhood pixels Sum of all coefficients are equal to 256, so "shift" value used is typically 8 to divide by 256. This parameter is configurable in case the user wants to multiply or divide the result by a power of 2. For example, if the user wants to multiply the result of the gaussian by 4, then the shift value should be 8-2 = 6.

Assumptions:

• I/O buffer pointers are assumed to be not aliased.
• Output height should be == (Input height - 4)
• Output width should be == (Input width - 4) OR (Input width)

Performance Considerations:

• For best performance, the following parameter settings are recommended:
  • Set widths equal each other and equal to strides (allows processing over whole image in single loop)

VXLIB_gaussian_5x5_i8u_o8u_checkParams()

VXLIB_STATUS VXLIB_gaussian_5x5_i8u_o8u_checkParams	(	const uint8_t	src[restrict],
		const VXLIB_bufParams2D_t *	src_addr,
		const uint8_t	dst[restrict],
		const VXLIB_bufParams2D_t *	dst_addr,
		uint8_t	shift
	)

Description:

Checks the parameters for programming errors for the VXLIB_gaussian_5x5_i8u_o8u function.

Method:

The following checks are made:

• There shall be no NULL pointers
• Output height should be == (Input height - 4)
• Output width should be <= Input width
• The strides of each image shall be equal to or greater than the x dimension
• shift should be <=15

Parameters

[in]	src[]	Pointer to array containing input image (UQ8.0)
[in]	src_addr[]	Pointer to structure containing dimensional information of src
[out]	dst[]	Pointer to array containing output image (UQ8.0)
[in]	dst_addr[]	Pointer to structure containing dimensional information of dst
[in]	shift	Right shift value after the sum of products of the 25 neighborhood pixels Sum of all coefficients are equal to 256, so "shift" value used is typically '8'. This parameter is configurable in case the user wants to multiply or divide the result by a power of 2. For example, if the user wants to multiply the result of the gaussian by 4, then the shift value should be 8-2 = 6.