VXLIB_sobel_3x3_i8u_o16s_o16s

Functions
VXLIB_STATUS	VXLIB_sobel_3x3_i8u_o16s_o16s (const uint8_t src[restrict], const VXLIB_bufParams2D_t *src_addr, int16_t dst_x[restrict], const VXLIB_bufParams2D_t *dst_x_addr, int16_t dst_y[restrict], const VXLIB_bufParams2D_t *dst_y_addr)
VXLIB_STATUS	VXLIB_sobel_3x3_i8u_o16s_o16s_checkParams (const uint8_t src[restrict], const VXLIB_bufParams2D_t *src_addr, const int16_t dst_x[restrict], const VXLIB_bufParams2D_t *dst_x_addr, const int16_t dst_y[restrict], const VXLIB_bufParams2D_t *dst_y_addr)

Detailed Description

Function Documentation

VXLIB_sobel_3x3_i8u_o16s_o16s()

VXLIB_STATUS VXLIB_sobel_3x3_i8u_o16s_o16s	(	const uint8_t	src[restrict],
		const VXLIB_bufParams2D_t *	src_addr,
		int16_t	dst_x[restrict],
		const VXLIB_bufParams2D_t *	dst_x_addr,
		int16_t	dst_y[restrict],
		const VXLIB_bufParams2D_t *	dst_y_addr
	)

Description:

Computes the sobel image filter kernel over a 3x3 window of imput image, for each pixel.

Method:

Produces 2 separate output planes in a single pass through the image; one plane for X and one plane for Y. The sobel operators are defined as:

       | -1 0 +1 |              | -1 -2 -1 |
 Gx =  | -2 0 +2 |        Gy =  |  0  0  0 |
       | -1 0 +1 |              | +1 +2 +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_x[]	Pointer to array containing X output image (SQ15.0)
[in]	dst_x_addr[]	Pointer to structure containing dimensional information of dst_x
[out]	dst_y[]	Pointer to array containing Y output image (SQ15.0)
[in]	dst_y_addr[]	Pointer to structure containing dimensional information of dst_y

Assumptions:

• I/O buffer pointers are assumed to be not aliased.
• Output height should be == (Input height - 2)
• Output width should be == (Input width - 2) OR (Input width)

Performance Considerations:

• For best performance, the following parameter settings are recommended:
  • Set output width values to a multiple of 8
  • Set widths equal to each other and equal to strides (allows processing over whole image in single loop)
  • If only x or y gradient is needed instead of both, one should use VXLIB_sobelX_3x3_i8u_o16s or VXLIB_sobelY_3x3_i8u_o16s, respecively.

VXLIB_sobel_3x3_i8u_o16s_o16s_checkParams()

VXLIB_STATUS VXLIB_sobel_3x3_i8u_o16s_o16s_checkParams	(	const uint8_t	src[restrict],
		const VXLIB_bufParams2D_t *	src_addr,
		const int16_t	dst_x[restrict],
		const VXLIB_bufParams2D_t *	dst_x_addr,
		const int16_t	dst_y[restrict],
		const VXLIB_bufParams2D_t *	dst_y_addr
	)

Description:

Checks the parameters for programming errors for the VXLIB_sobel_3x3_i8u_o16s_o16s function.

Method:

The following checks are made:

• There shall be no NULL pointers
• Output height should be == (Input height - 2)
• Output width should be <= Input width
• The strides of each image shall be equal to or greater than the x dimension

Parameters

[in]	src[]	Pointer to array containing input image (UQ8.0)
[in]	src_addr[]	Pointer to structure containing dimensional information of src
[out]	dst_x[]	Pointer to array containing X output image (SQ15.0)
[in]	dst_x_addr[]	Pointer to structure containing dimensional information of dst_x
[out]	dst_y[]	Pointer to array containing Y output image (SQ15.0)
[in]	dst_y_addr[]	Pointer to structure containing dimensional information of dst_y