Debian Codegraph @

find_board.docstring

raw
1 Runs the full mrgingham chessboard detector on an image
2
3 SYNOPSIS
4
5 points = mrgingham.find_board(image)
6
7 [ points is an array of shape (N*N,2), the ordered list of ]
8 [ pixel coordinates in the grid ]
9
10 This function runs the full mrgingham sequence:
11
12 - pre-process image
13 - find chessboard corners (or blobs)
14 - find an NxN grid in the set of corners
15 - refine
16
17 No broadcasting is supported by this function
18
19 ARGUMENTS
20
21 - image: numpy array of shape (H,W) and dtype np.uint8. This is the image we're
22 processing. Must be densely-stored, grayscale image
23
24 - blobs: optional boolean, defaulting to False. If True, we look for
25 black-on-white blobs, instead of chessboard corners. If blobs: we MUST have
26 image_pyramid_level==0
27
28 - image_pyramid_level: optional integer defaulting to -1. This can be given to
29 operate on a downsampled version of the image. 0 means "original image", 1
30 means "downsample by a factor of 2 in each dimension", 2 means "downsample by
31 a factor of 4 in each dimension" and so on. image_pyramid_level < 0 (the
32 default) means "start with a downsampled image, and then refine the results by
33 repeatedly reducing the downsampling. If blobs: image_pyramid_level==0 is the
34 only allowed option
35
36 - gridn: optional integer defaulting to 10. We detect an NxN grid of corners,
37 where N is given by this argument
38
39 - debug: optional boolean, defaulting to False. If True, this will dump various
40 intermediate results into /tmp and it will report more stuff on standard out.
41 Most of the intermediate results are self-plotting data files. Run them. This
42 does the same thing as "mrgingham --debug"; the Python output is unaffected.
43
44 - debug_sequence: optional string "X,Y" where X and Y are integers. In addition
45 to "debug", this produces diagnostics when searching for sequence candidates.
46 X,Y are the approximate image coordinates of the start of a given sequence
47 (corner on the edge of a chessboard. This doesn't need to be exact; mrgingham
48 will report on the nearest corner. This does the same thing as "mrgingham
49 --debug-sequence"; the Python output is unaffected.
50
51 RETURNED VALUE
52
53 A numpy array of shape (N*N,2) containing ordered pixel coordinates of the grid
54 found in the image. If no grid was found, None is returned.