morph: Morphological Operations
In neuroconductor-devel/Rvision: Basic Computer Vision Library

morph applies various morphological operations (see Note) to an Image object.

1 2	morph(image, operation, kernel = NULL, k_shape = "rectangle", k_height = 5, k_width = 5, iterations = 1)

`image`	An `Image` object.
`operation`	A string corresponding to the name of a morphological operation to apply to the image (see Note).
`kernel`	A binary matrix. If the matrix is not binary, all positive values will be automatically converted to 1, all negative values to 0, unless `operation = "hitmiss"` in which case they will be converted to -1.
`k_shape`	A string corresponding to the shape of the kernel for the morphological operation (see Note; default: "rectangle"). Ignored if a custom `kernel` is provided.
`k_height`	The half-height in pixels of the kernel. Ignored if a custom `kernel` is provided.
`k_width`	The half-width in pixels of the kernel. Ignored if a custom `kernel` is provided.
`iterations`	The number of times the morphological operations should be applied.

An Image object.

There are 8 types of morphological operations that can be achieved with this function:

"erode":for each point, returns the minimum of the points in its neighborhood, with that neighborhood defined by the kernel.
"dilate":for each point, returns the maximum of the points in its neighborhood, with that neighborhood defined by the kernel.
"open":erosion followed by dilation.
"close":dilation followed by erosion.
"gradient":difference between the dilation and the erosion of an image.
"tophat":difference between an input image and its opening.
"blackhat":difference between the closing and its input image.
"hitmiss":(1) erodes the image with kernel > 0; (2) erodes the complement of the image with kernel < 0; (3) returns the intersection (AND) of step 1 and step 2. The hit-or-miss transform is the basis of more advanced morphological operations such as thinning or pruning.