The OpenVX Specification  a73e458
Canny Edge Detector

Detailed Description

Provides a Canny edge detector kernel.

This function implements an edge detection algorithm similar to that described in [2]. The main components of the algorithm are:

The details of each of these steps are described below.

\[ \mathbf{sobel}_{y}=transpose({sobel}_{x})= \begin{vmatrix} -1 & -2 & -1\\ 0 & 0 & 0\\ 1 & 2 & 1 \end{vmatrix} \]


 Normalization type constants.


vx_node VX_API_CALL vxCannyEdgeDetectorNode (vx_graph graph, vx_image input, vx_threshold hyst, vx_int32 gradient_size, vx_enum norm_type, vx_image output)
 [Graph] Creates a Canny Edge Detection Node. More...

Function Documentation

◆ vxCannyEdgeDetectorNode()

vx_node VX_API_CALL vxCannyEdgeDetectorNode ( vx_graph  graph,
vx_image  input,
vx_threshold  hyst,
vx_int32  gradient_size,
vx_enum  norm_type,
vx_image  output 

[Graph] Creates a Canny Edge Detection Node.

[in]graphThe reference to the graph [R00338].
[in]inputThe input VX_DF_IMAGE_U8 image [R00339].
[in]hystThe double threshold for hysteresis [R00340]. The threshold data_type shall be either VX_TYPE_UINT8 or VX_TYPE_INT16 [R00341]. The VX_THRESHOLD_TRUE_VALUE and VX_THRESHOLD_FALSE_VALUE of vx_threshold are ignored.
[in]gradient_sizeThe size of the Sobel filter window, must support at least 3, 5, and 7 [R00342].
[in]norm_typeA flag indicating the norm used to compute the gradient, VX_NORM_L1 or VX_NORM_L2 [R00343].
[out]outputThe output image in VX_DF_IMAGE_U8 format with values either 0 or 255 [R00344].
Return values
vx_nodeA node reference. Any possible errors preventing a successful creation should be checked using vxGetStatus