skimage2.morphology.area_closing#
- skimage2.morphology.area_closing(image, area_threshold=64, connectivity=1, parent=None, tree_traverser=None)[source]#
Perform an area closing of the image.
Area closing removes all dark structures of an image with a surface smaller than area_threshold. The output image is larger than or equal to the input image for every pixel and all local minima have at least a surface of area_threshold pixels.
Area closings are similar to morphological closings, but they do not use a fixed footprint, but rather a deformable one, with surface = area_threshold.
In the binary case, area closings are equivalent to remove_small_holes; this operator is thus extended to gray-level images.
Technically, this operator is based on the max-tree representation of the image.
- Parameters:
- imagendarray
The input image for which the area_closing is to be calculated. This image can be of any type.
- area_thresholdunsigned int
The size parameter (number of pixels). The default value is arbitrarily chosen to be 64.
- connectivityunsigned int, optional
The neighborhood connectivity. The integer represents the maximum number of orthogonal steps to reach a neighbor. In 2D, it is 1 for a 4-neighborhood and 2 for a 8-neighborhood. Default value is 1.
- parentndarray, int64, optional
Parent image representing the max tree of the inverted image. The value of each pixel is the index of its parent in the ravelled array. See Note for further details.
- tree_traverser1D array, int64, optional
The ordered pixel indices (referring to the ravelled array). The pixels are ordered such that every pixel is preceded by its parent (except for the root which has no parent).
- Returns:
- outputndarray
Output image of the same shape and type as input image.
See also
Notes
If a max-tree representation (parent and tree_traverser) are given to the function, they must be calculated from the inverted image for this function, i.e.: >>> from _skimage2.morphology import diameter_closing, max_tree >>> from _skimage2.util import invert >>> P, S = max_tree(invert(f)) >>> closed = diameter_closing(f, 3, parent=P, tree_traverser=S)
References
[1]Vincent L., Proc. “Grayscale area openings and closings, their efficient implementation and applications”, EURASIP Workshop on Mathematical Morphology and its Applications to Signal Processing, Barcelona, Spain, pp.22-27, May 1993.
[2]Soille, P., “Morphological Image Analysis: Principles and Applications” (Chapter 6), 2nd edition (2003), ISBN 3540429883. DOI:10.1007/978-3-662-05088-0
[3]Salembier, P., Oliveras, A., & Garrido, L. (1998). Antiextensive Connected Operators for Image and Sequence Processing. IEEE Transactions on Image Processing, 7(4), 555-570. DOI:10.1109/83.663500
[4]Najman, L., & Couprie, M. (2006). Building the component tree in quasi-linear time. IEEE Transactions on Image Processing, 15(11), 3531-3539. DOI:10.1109/TIP.2006.877518
[5]Carlinet, E., & Geraud, T. (2014). A Comparative Review of Component Tree Computation Algorithms. IEEE Transactions on Image Processing, 23(9), 3885-3895. DOI:10.1109/TIP.2014.2336551
Examples
We create an image (quadratic function with a minimum in the center and 4 additional local minima.
>>> w = 12 >>> x, y = np.mgrid[0:w,0:w] >>> f = 180 + 0.2*((x - w/2)**2 + (y-w/2)**2) >>> f[2:3,1:5] = 160; f[2:4,9:11] = 140; f[9:11,2:4] = 120 >>> f[9:10,9:11] = 100; f[10,10] = 100 >>> f = f.astype(int)
We can calculate the area closing:
>>> closed = area_closing(f, 8, connectivity=1)
All small minima are removed, and the remaining minima have at least a size of 8.