harrisCorners

Calculate per-pixel corner impuls response using Harris corner detector.

Slice!(SliceKind.contiguous, [2], OutputType*)

harrisCorners

(

InputType

OutputType = InputType

SliceKind inputKind

)

(

Slice!(inputKind, [2], InputType*) image

in uint winSize = 3

in float k = 0.64f

in float gauss = 0.84f

Slice!(SliceKind.contiguous, [2], OutputType*) prealloc = emptySlice!([2], OutputType)

TaskPool pool = taskPool

)

in { assert (!image.empty, "Empty image given."); assert (winSize % 2 != 0, "Kernel window size has to be odd."); assert (gauss > 0.0, "Gaussian sigma value has to be greater than 0."); assert (k > 0.0, "K value has to be greater than 0."); }

Parameters

image: Type: Slice!(inputKind, [2], InputType*)
Input image slice.
winSize: Type: uint
Window (square) size used in corner detection.
k: Type: float
Sensitivity parameter defined in the algorithm.
gauss: Type: float
Gauss sigma value used as window weighting parameter.
prealloc: Type: Slice!(SliceKind.contiguous, [2], OutputType*)
Optional pre-allocated buffer for return response image.
pool: Type: TaskPool
TaskPool instance used parallelise the algorithm.

Return Value

Type: Slice!(SliceKind.contiguous, [2], OutputType*)

Response matrix the same size of the input image, where each pixel represents corner response value - the bigger the value, more probably it represents the actual corner in the image.

harrisCorners

Parameters

Return Value

Meta

Source