9i)rSrSSKrSSKJr SSKJrJr SSKJ r J r J r SSK J r SSSS S .S jrSSSS .S jrg)zflood_fill.py - in place flood fill algorithm This module provides a function to fill all equal (or within tolerance) values connected to a given seed point with a different value. N)crop)_flood_fill_equal_flood_fill_tolerance)_offsets_to_raveled_neighbors_resolve_neighborhood_set_border_values)numeric_dtype_min_maxF) footprint connectivity tolerancein_placecV[UUUUUS9nU(dUR5nX U'U$)at Perform flood filling on an image. Starting at a specific `seed_point`, connected points equal or within `tolerance` of the seed value are found, then set to `new_value`. Parameters ---------- image : ndarray An n-dimensional array. seed_point : tuple or int The point in `image` used as the starting point for the flood fill. If the image is 1D, this point may be given as an integer. new_value : `image` type New value to set the entire fill. This must be chosen in agreement with the dtype of `image`. footprint : ndarray, optional The footprint (structuring element) used to determine the neighborhood of each evaluated pixel. It must contain only 1's and 0's, have the same number of dimensions as `image`. If not given, all adjacent pixels are considered as part of the neighborhood (fully connected). connectivity : int, optional A number used to determine the neighborhood of each evaluated pixel. Adjacent pixels whose squared distance from the center is less than or equal to `connectivity` are considered neighbors. Ignored if `footprint` is not None. tolerance : float or int, optional If None (default), adjacent values must be strictly equal to the value of `image` at `seed_point` to be filled. This is fastest. If a tolerance is provided, adjacent points with values within plus or minus tolerance from the seed point are filled (inclusive). in_place : bool, optional If True, flood filling is applied to `image` in place. If False, the flood filled result is returned without modifying the input `image` (default). Returns ------- filled : ndarray An array with the same shape as `image` is returned, with values in areas connected to and equal (or within tolerance of) the seed point replaced with `new_value`. Notes ----- The conceptual analogy of this operation is the 'paint bucket' tool in many raster graphics programs. Examples -------- >>> from skimage.morphology import flood_fill >>> image = np.zeros((4, 7), dtype=int) >>> image[1:3, 1:3] = 1 >>> image[3, 0] = 1 >>> image[1:3, 4:6] = 2 >>> image[3, 6] = 3 >>> image array([[0, 0, 0, 0, 0, 0, 0], [0, 1, 1, 0, 2, 2, 0], [0, 1, 1, 0, 2, 2, 0], [1, 0, 0, 0, 0, 0, 3]]) Fill connected ones with 5, with full connectivity (diagonals included): >>> flood_fill(image, (1, 1), 5) array([[0, 0, 0, 0, 0, 0, 0], [0, 5, 5, 0, 2, 2, 0], [0, 5, 5, 0, 2, 2, 0], [5, 0, 0, 0, 0, 0, 3]]) Fill connected ones with 5, excluding diagonal points (connectivity 1): >>> flood_fill(image, (1, 1), 5, connectivity=1) array([[0, 0, 0, 0, 0, 0, 0], [0, 5, 5, 0, 2, 2, 0], [0, 5, 5, 0, 2, 2, 0], [1, 0, 0, 0, 0, 0, 3]]) Fill with a tolerance: >>> flood_fill(image, (0, 0), 5, tolerance=1) array([[5, 5, 5, 5, 5, 5, 5], [5, 5, 5, 5, 2, 2, 5], [5, 5, 5, 5, 2, 2, 5], [5, 5, 5, 5, 5, 5, 3]]) r r r)floodcopy)image seed_point new_valuer r rrmasks ^/var/www/html/land-doc-ocr/venv/lib/python3.13/site-packages/skimage/morphology/_flood_fill.py flood_fillrs:~  !  D  $K Lrc [R"U5nURRSLaSnO4URRSLaSnO[R "U5nSnSUR ;a#[R"UR [S9$[U5 Xn[[R"U5UR -5n[X#URSS9n[SUR 55n[[R"U5U5VV s/sH5up[R "[R""X- 554S -PM7 n nn [R$"X S UR'5S 9n [R("[X5V V Vs/sH un upX-PM snn n U R US 9n[+U R X'US 9n[R"U R [R,US9n[/US U S9 Ub[#U5n[1UR25unn[!UR55UR55U- 5n['UR55UR55U-5n[7U R9U5UR9U5UUUUU5 O.[;U R9U5UR9U5UUU5 [?UU SS9RA[5$![a U4nGNf=fs sn nfs snn n f![a+ U R2[R<:Xa [S5eef=f)a Mask corresponding to a flood fill. Starting at a specific `seed_point`, connected points equal or within `tolerance` of the seed value are found. Parameters ---------- image : ndarray An n-dimensional array. seed_point : tuple or int The point in `image` used as the starting point for the flood fill. If the image is 1D, this point may be given as an integer. footprint : ndarray, optional The footprint (structuring element) used to determine the neighborhood of each evaluated pixel. It must contain only 1's and 0's, have the same number of dimensions as `image`. If not given, all adjacent pixels are considered as part of the neighborhood (fully connected). connectivity : int, optional A number used to determine the neighborhood of each evaluated pixel. Adjacent pixels whose squared distance from the center is less than or equal to `connectivity` are considered neighbors. Ignored if `footprint` is not None. tolerance : float or int, optional If None (default), adjacent values must be strictly equal to the initial value of `image` at `seed_point`. This is fastest. If a value is given, a comparison will be done at every point and if within tolerance of the initial value will also be filled (inclusive). Returns ------- mask : ndarray A Boolean array with the same shape as `image` is returned, with True values for areas connected to and equal (or within tolerance of) the seed point. All other values are False. Notes ----- The conceptual analogy of this operation is the 'paint bucket' tool in many raster graphics programs. This function returns just the mask representing the fill. If indices are desired rather than masks for memory reasons, the user can simply run `numpy.nonzero` on the result, save the indices, and discard this mask. Examples -------- >>> from skimage.morphology import flood >>> image = np.zeros((4, 7), dtype=int) >>> image[1:3, 1:3] = 1 >>> image[3, 0] = 1 >>> image[1:3, 4:6] = 2 >>> image[3, 6] = 3 >>> image array([[0, 0, 0, 0, 0, 0, 0], [0, 1, 1, 0, 2, 2, 0], [0, 1, 1, 0, 2, 2, 0], [1, 0, 0, 0, 0, 0, 3]]) Fill connected ones with 5, with full connectivity (diagonals included): >>> mask = flood(image, (1, 1)) >>> image_flooded = image.copy() >>> image_flooded[mask] = 5 >>> image_flooded array([[0, 0, 0, 0, 0, 0, 0], [0, 5, 5, 0, 2, 2, 0], [0, 5, 5, 0, 2, 2, 0], [5, 0, 0, 0, 0, 0, 3]]) Fill connected ones with 5, excluding diagonal points (connectivity 1): >>> mask = flood(image, (1, 1), connectivity=1) >>> image_flooded = image.copy() >>> image_flooded[mask] = 5 >>> image_flooded array([[0, 0, 0, 0, 0, 0, 0], [0, 5, 5, 0, 2, 2, 0], [0, 5, 5, 0, 2, 2, 0], [1, 0, 0, 0, 0, 0, 3]]) Fill with a tolerance: >>> mask = flood(image, (0, 0), tolerance=1) >>> image_flooded = image.copy() >>> image_flooded[mask] = 5 >>> image_flooded array([[5, 5, 5, 5, 5, 5, 5], [5, 5, 5, 5, 2, 2, 5], [5, 5, 5, 5, 2, 2, 5], [5, 5, 5, 5, 5, 5, 3]]) TFCr)dtypeF)enforce_adjacencyc3*# UH oS-v M g7f)rN).0ss r flood..s3?a6?srconstant)modeconstant_values)order)centerr))rr))value border_widthzLdtype of `image` is float16 which is not supported, try upcasting to float32)r)!npasarrayflags f_contiguous c_contiguousascontiguousarrayshapezerosbooliter TypeErrortupler ndimzipnonzeromaxabspadminravel_multi_indexruint8r r ritemrravelrfloat16rview)rrr r rr) seed_valuer*idxc pad_width working_imagei pad_startpad_endravelled_seed_idxneighbor_offsetsr/ min_value max_valuelow_tolhigh_tols rrrs| JJu E {{4'  ! !T )$$U+ EKKxx 400# Z"Jrzz*- ;"  II$99I9I#J Iy)..*JOO,= ,IJG9>>+Z__->-JKH !##E* E" !  ##E* E" !  $ yu - 2 24 88O # ] # RL    "** ,6  s1 L*rZsH D 2 k\+/TTu9r