vpi.Image.harriscorners

Image.harriscorners(self: vpi.Image, *, backend: vpi.Backend = vpi.Backend.DEFAULT, out_features: vpi.Array = None, out_scores: vpi.Array = None, gradient_size: int = 3, block_size: int = 3, strength: float = 20.0, sensitivity: float = 0.0625, min_nms_distance: float = 8.0, stream: vpi.Stream = None) → Tuple[vpi.Array, vpi.Array]

Runs Harris corner detector on the image.

See also

Refer to the algorithm explanation for more details and usage examples.

Parameters

  • out_features (vpi.Array, optional) – Array that will receive the detected corners. Array size is updated with the number of corners found.

  • out_scores (vpi.Array, optional) – Array that will receive the corners’ scores. Array size matches outFeatures array’ size.

  • gradient_size (int, optional) – Gradient window size.

  • block_size (int, optional) – Block window size used to compute the Harris Corner score.

  • strength (float, optional) – Specifies the minimum threshold with which to eliminate Harris Corner scores.

  • sensitivity (float, optional) – Specifies sensitivity threshold from the Harris-Stephens equation.

  • min_nms_distance (float, optional) – Non-maximum suppression radius, set to 0 to disable it.

  • out (vpi.Image, optional) – The output for the algorithm.

  • backend (vpi.Backend, optional) – The backend to be used by the algorithm.

  • stream (vpi.Stream, optional) – The stream to be used by the algorithm.

Returns

A tuple with the output features and scores array for the algorithm, respectively.

Return type

Tuple[vpi.Array, vpi.Array]

Caution

Restrictions to several arguments may apply. Check the C API references of the submit function and the group concepts for more details.