Hough Lines

Detailed Description

The Hough Lines primitive finds lines in a binary image using Hough transform.

The primitive uses the following identifiers:

• NVX_KERNEL_HOUGH_LINES and NVX_KERNEL_HOUGH_SEGMENTS for vxGetKernelByEnum function;
• "com.nvidia.nvx.hough_lines" and "com.nvidia.nvx.hough_segments" for vxGetKernelByName function.

Method

The primitive has these two modes:

• Standard—Standard Hough transform is explained here: [10]

  Each line is represented by a two-element vector \( (\rho, \theta) \).

  \( \rho \) is the distance from the coordinate origin. \( (0, 0) \) (top-left corner of the image).

  \( \theta \) is the line rotation angle in radians. \( (0 \sim \textrm{vertical line}, \pi/2 \sim \textrm{horizontal line}) \).

• Probabilistic—Probabilistic Hough transform is described in [5].

  Probabilistic Hough transform is more efficient if the picture contains a few long linear segments. It returns line segments rather than the whole line.

  Each line is represented by a 4-element vector \( (x_1, y_1, x_2, y_2) \), where \( (x_1, y_1) \) and \( (x_2, y_2) \). are the ending points of each detected line segment.

Functions
vx_node	nvxHoughLinesNode (vx_graph graph, vx_image input, vx_array output, vx_float32 rho, vx_float32 theta, vx_uint32 threshold, vx_scalar num_lines)
	[Graph] Finds lines on image using standard Hough transform. More...
vx_node	nvxHoughSegmentsNode (vx_graph graph, vx_image input, vx_array output, vx_float32 rho, vx_float32 theta, vx_uint32 threshold, vx_uint32 minLineLength, vx_uint32 maxLineGap, vx_scalar num_segments)
	[Graph] Finds line segments in a binary image using the probabilistic Hough transform. More...
vx_status	nvxuHoughLines (vx_context context, vx_image input, vx_array output, vx_float32 rho, vx_float32 theta, vx_uint32 threshold, vx_scalar num_lines)
	[Immediate] Finds lines on image using standard Hough transform. More...
vx_status	nvxuHoughSegments (vx_context context, vx_image input, vx_array output, vx_float32 rho, vx_float32 theta, vx_uint32 threshold, vx_uint32 minLineLength, vx_uint32 maxLineGap, vx_scalar num_segments)
	[Immediate] Finds line segments in a binary image using the probabilistic Hough transform. More...

Function Documentation

vx_node nvxHoughLinesNode	(	vx_graph	graph,
		vx_image	input,
		vx_array	output,
		vx_float32	rho,
		vx_float32	theta,
		vx_uint32	threshold,
		vx_scalar	num_lines
	)

[Graph] Finds lines on image using standard Hough transform.

Parameters

[in]	graph	Specifies the graph.
[in]	input	Specifies the input image. Only VX_DF_IMAGE_U8 format is supported. The image width and height must be less than \( 2^{16} \).
[out]	output	Specifies the output lines list. Only NVX_TYPE_POINT3F item type is supported. The array capacity must be explicitly provided, even for virtual arrays.
[in]	rho	Specifies the distance resolution of the accumulator in pixels. It must be a positive value.
[in]	theta	Specifies the angle resolution of the accumulator in radians. It must be in range \( (0, \pi] \).
[in]	threshold	Specifies the accumulator threshold parameter. Only lines that get enough votes (> threshold) are returned.
[out]	num_lines	Specifies the total number of detected lines in image (optional). It must be a VX_TYPE_UINT32 scalar.

Returns

A valid node reference or an error object (use vxGetStatus).

See also

Hough Lines

vx_node nvxHoughSegmentsNode	(	vx_graph	graph,
		vx_image	input,
		vx_array	output,
		vx_float32	rho,
		vx_float32	theta,
		vx_uint32	threshold,
		vx_uint32	minLineLength,
		vx_uint32	maxLineGap,
		vx_scalar	num_segments
	)

[Graph] Finds line segments in a binary image using the probabilistic Hough transform.

Parameters

[in]	graph	Specifies the graph.
[in]	input	Specifies the input image (8-bit grayscale). Only VX_DF_IMAGE_U8 format is supported. The image width and height must be less than \( 2^{16} \).
[out]	output	Specifies the output lines list. Only NVX_TYPE_POINT4F item type is supported. The array capacity must be explicitly provided, even for virtual arrays.
[in]	rho	Specifies the distance resolution of the accumulator in pixels. It must be a positive value.
[in]	theta	Specifies the angle resolution of the accumulator in radians. It must be in range \( (0, \pi] \).
[in]	threshold	Specifies the accumulator threshold parameter. Only lines that get enough votes (> threshold) are returned.
[in]	minLineLength	Specifies the minimum line length. Shorter segments are rejected.
[in]	maxLineGap	Specifies the maximum allowed gap between points on the same line to link them.
[out]	num_segments	Specifies the total number of detected segments in image (optional). It should be VX_TYPE_UINT32 scalar.

Returns

A valid node reference or an error object (use vxGetStatus).

See also

Hough Lines

vx_status nvxuHoughLines	(	vx_context	context,
		vx_image	input,
		vx_array	output,
		vx_float32	rho,
		vx_float32	theta,
		vx_uint32	threshold,
		vx_scalar	num_lines
	)

[Immediate] Finds lines on image using standard Hough transform.

Parameters

[in]	context	Specifies the context.
[in]	input	Specifies the input image (8-bit grayscale). Only VX_DF_IMAGE_U8 format is supported. The image width and height must be less than \( 2^{16} \).
[out]	output	Specifies the output lines list. Only NVX_TYPE_POINT3F item type is supported. The array capacity must be explicitly provided, even for virtual arrays.
[in]	rho	Specifies the distance resolution of the accumulator in pixels. It must be a positive value.
[in]	theta	Specifies the angle resolution of the accumulator in radians. It must be in range \( (0, \pi] \).
[in]	threshold	Specifies the accumulator threshold parameter. Only those lines are returned that get enough votes (> threshold).
[out]	num_lines	Specifies the total number of detected lines in image (optional). It should be VX_TYPE_UINT32 scalar.

Returns

A vx_status enumerator.

Return values

VX_SUCCESS	No errors.
VX_ERROR_INVALID_REFERENCE	Supplied parameters are not a valid references.
VX_ERROR_INVALID_PARAMETERS	Supplied parameters are not valid.
VX_ERROR_INVALID_SCOPE	Supplied parameters are virtual objects, that cannot be used in immediate mode.
VX_FAILURE	Internal error in primitive implementation; check the log for detailed information (Framework: Log).

See also

Hough Lines

vx_status nvxuHoughSegments	(	vx_context	context,
		vx_image	input,
		vx_array	output,
		vx_float32	rho,
		vx_float32	theta,
		vx_uint32	threshold,
		vx_uint32	minLineLength,
		vx_uint32	maxLineGap,
		vx_scalar	num_segments
	)

[Immediate] Finds line segments in a binary image using the probabilistic Hough transform.

Parameters

[in]	context	Specifies the context.
[in]	input	Specifies the input image (8-bit grayscale). Only VX_DF_IMAGE_U8 format is supported. The image width and height must be less than \( 2^{16} \).
[out]	output	Specifies the output lines list. Only NVX_TYPE_POINT4F item type is supported. The array capacity must be explicitly provided, even for virtual arrays.
[in]	rho	Specifies the distance resolution of the accumulator in pixels. It must be a positive value.
[in]	theta	Specifies the angle resolution of the accumulator in radians. It must be in range \( (0, \pi] \).
[in]	threshold	Specifies the accumulator threshold parameter. Only lines that get enough votes (> threshold) are returned.
[in]	minLineLength	Specifies the minimum line length. Shorter segments are rejected.
[in]	maxLineGap	Specifies the maximum allowed gap between points on the same line to link them.
[out]	num_segments	Specifies the total number of detected segments in image (optional). It should be VX_TYPE_UINT32 scalar.

Returns

A vx_status enumerator.

Return values

VX_SUCCESS	No errors.
VX_ERROR_INVALID_REFERENCE	Supplied parameters are not a valid references.
VX_ERROR_INVALID_PARAMETERS	Supplied parameters are not valid.
VX_ERROR_INVALID_SCOPE	Supplied parameters are virtual objects, that cannot be used in immediate mode.
VX_FAILURE	Internal error in primitive implementation, check log for detailed information (Framework: Log).

See also

Hough Lines