Marching squares: Difference between revisions
m
→{{header|11l}}: X.throw
m (→{{header|J}}: not quite so lame, but still lame) |
Alextretyak (talk | contribs) m (→{{header|11l}}: X.throw) |
||
| (12 intermediate revisions by 7 users not shown) | |||
Line 6:
See: [https://en.wikipedia.org/wiki/Marching_squares Marching squares]
=={{header|11l}}==
{{trans|Wren}}
<syntaxhighlight lang="11l">
-V
DIR_E = ( 1, 0)
DIR_NE = ( 1, 1)
DIR_N = ( 0, 1)
DIR_NW = (-1, 1)
DIR_W = (-1, 0)
DIR_SW = (-1, -1)
DIR_S = ( 0, -1)
DIR_SE = ( 1, -1)
F path_str(origin_x, origin_y, directions)
-V :dirn = [:DIR_E = ‘E’,
:DIR_NE = ‘NE’,
:DIR_N = ‘N’,
:DIR_NW = ‘NW’,
:DIR_W = ‘W’,
:DIR_SW = ‘SW’,
:DIR_S = ‘S’,
:DIR_SE = ‘SE’]
R ‘X: ’origin_x‘, Y: ’origin_y‘, Path: ’directions.map(d -> @:dirn[d])
T MarchingSquares
. Int width, height
. [Int] data
F (width, height, data)
.width = width
.height = height
.data = copy(data)
. F is_set(x, y)
I x <= 0 | x > .width | y <= 0 | y > .height
R 0B
R .data[(y - 1) * .width + (x - 1)] != 0
. F value(x, y)
V sum = 0
I .is_set(x, y) {sum [|]= 1}
I .is_set(x + 1, y) {sum [|]= 2}
I .is_set(x, y + 1) {sum [|]= 4}
I .is_set(x + 1, y + 1) {sum [|]= 8}
R sum
. F identify_perimeter_(=initial_x, =initial_y)
I initial_x < 0 {initial_x = 0}
I initial_x > .width {initial_x = .width}
I initial_y < 0 {initial_y = 0}
I initial_y > .height {initial_y = .height}
V initial_value = .value(initial_x, initial_y)
I initial_value C (0, 15)
X.throw RuntimeError(‘Supplied initial coordinates (#., #.) do not lie on a perimeter.’.format(initial_x, initial_y))
[IVec2] directions
V x = initial_x
V y = initial_y
V previous = (0, 0)
L
IVec2 direction
S .value(x, y)
1, 5, 13
direction = :DIR_N
2, 3, 7
direction = :DIR_E
4, 12, 14
direction = :DIR_W
8, 10, 11
direction = :DIR_S
6
direction = I previous == :DIR_N {:DIR_W} E :DIR_E
9
direction = I previous == :DIR_E {:DIR_N} E :DIR_S
E
X.throw RuntimeError(‘Illegal state.’)
directions.append(direction)
x += direction.x
y -= direction.y
previous = direction
I x == initial_x & y == initial_y
L.break
R path_str(initial_x, -initial_y, directions)
F identify_perimeter()
V size = .width * .height
L(i) 0 .< size
I .data[i] != 0
R .identify_perimeter_(i % .width, i I/ .width)
V example = [
0, 0, 0, 0, 0,
0, 0, 0, 0, 0,
0, 0, 1, 1, 0,
0, 0, 1, 1, 0,
0, 0, 0, 1, 0,
0, 0, 0, 0, 0
]
V ms = MarchingSquares(5, 6, example)
print(ms.identify_perimeter())
</syntaxhighlight>
{{out}}
<pre>
X: 2, Y: -2, Path: [S, S, E, S, E, N, N, N, W, W]
</pre>
=={{header|J}}==
Line 11 ⟶ 123:
This is a partial implementation, see the talk page for some discussion of the untouched issues.
<
step2=: {{(($y)#:i.$y) {{<x step2a y{::LUT}}"1 0 y}}
step2a=: {{ if. #y do. x+"1 y else. y end. }}
Line 49 ⟶ 161:
end.
r,<~.c
}}</
Task example:
<
img
1 1
Line 69 ⟶ 181:
│2 4│
│1 3│
└───┘</
Here, <code>img</code> is a bitmap. We pad the bitmap by surrounding it with zeros during processing. The box at the end contains a contour corresponding to the bitmap. Here, the first column represents row number (row 0 at the top) and the second column represents column number (column 0 at the left). Each contour represents a closed loop (so the final coordinate pair would connect with the first coordinate pair).
While the above implementation is incomplete, it seems to adequately handle an oval of cassini with focal points at X=1, -1 and Y=0:
<syntaxhighlight lang="j">a=: 1
X=: |:Y=:201#"0]0.02*i:100
Z0=: (*:(*:X)+(*:Y)) + (_2*(*:a)*X -&*: Y) + *:a
Z=: (Z0>:0.8)*Z0<:1.2
C=: unwind step2 step1 Z</syntaxhighlight>
Here, Z is a bitmap, and C is a list of three contours (one with 336 distinct coordinate pairs, and two with 134 distinct coordinate pairs) which surround that bitmap. These can be inspected (after <code>require'viewmat'</code>) with <code>viewmat Z</code> and <code>viewmat 1 (<"1;C)} 200 200$0</code>, and look plausible.
(Presumably the above implementation would fail if a threshold had been picked such that the bitmap exhibited a saddlepoint at the origin.)
=={{header|Julia}}==
Uses the marching squares algorithm: see github.com/JuliaGeometry/Contour.jl/blob/master/src/Contour.jl
See the discussion page for the Oval of Cassini example
<syntaxhighlight lang="julia">using Contour
import GLMakie as GM # GLMakie also defines Contour so import, not using
const example = Float64.([
Line 86 ⟶ 212:
])
const cl = first(levels(contours(1:6, 1:5, example)))
const xs, ys = coordinates(first(lines(cl)))
# Showing the points of the contour as origin (0, 0) at bottom left
const points = [(Int(round(ys[i])) - 1, 6 - Int(round(xs[i]))) for i in eachindex(xs)]
@show points
# oval of Cassini formula in z below, see formula at en.wikipedia.org/wiki/Cassini_oval#Equations
const xarray, yarray, a = -2.0:0.02:2.0, -2.0:0.02:2.0, 1.0
const z = [isapprox((x^2 + y^2)^2 - 2 * a^2 * (x^2 - y^2) + a^2, 1.0, atol=0.2) ? 1.0 : 0.0
for x in xarray, y in yarray]
# The first (and pehaps only significant) level is the 0 <-> 1 transition border
# There are 3 separate contours at that level, on outside and 2 holes
const figeight = levels(contours(1:size(z, 1), 1:size(z, 2), z))
const ovalxs, ovalys = coordinates(lines(figeight[1])[1])
const ovalxs2, ovalys2 = coordinates(lines(figeight[1])[2])
const ovalxs3, ovalys3 = coordinates(lines(figeight[1])[3])
const oplot = GM.plot(z)
GM.lines!(ovalxs, ovalys, color = :red, linewidth = 4)
GM.lines!(ovalxs2, ovalys2, color = :white, linewidth = 4)
GM.lines!(ovalxs3, ovalys3, color = :lightgreen, linewidth = 4)
GM.display(oplot)
</syntaxhighlight>{{out}}
<pre>
points = [(3, 4), (4, 3), (4, 2), (4, 1), (3, 0), (2, 1), (2, 1), (1, 2), (1, 3), (2, 4), (3, 4)]
</pre>
=={{header|Lua}}==
Based on the Phix and Wren solutions.
<
local Directions = { -- clockwise from North
N = {x= 0, y=-1},
Line 205 ⟶ 339:
msMain (1, example)
</
<pre>
root: x=1 y=2
Line 212 ⟶ 346:
positions: {1,2, 2,2, 2,3, 3,3, 4,3, 4,4, 5,4, 5,3, 6,3, 6,2, 7,2, 7,3, 6,3, 6,4, 6,5, 6,6, 7,6, 7,7, 6,7, 6,6, 5,6, 4,6, 4,5, 3,5, 3,6, 2,6, 2,7, 1,7, 1,6, 2,6, 2,5, 2,4, 2,3, 1,3, }
</pre>
=={{header|Perl}}==
{{trans|Raku}}
<syntaxhighlight lang="perl">use v5.36;
no warnings 'experimental::for_list';
use List::Util 'any';
use enum <E N W S>;
sub X ($a,$b) { my @c; for my $aa (0..$a) { for my $bb (0..$b) { push @c, $aa, $bb } } @c }
sub identify_perimeter(@data) {
for my ($x,$y) (X $#{$data[0]}, $#data) {
next unless $data[$y][$x] and $data[$y][$x] != 0;
my ($path,$cx,$cy,$d,$p) = ('', $x, $y);
do {
my $mask;
for my($dx,$dy,$b) (0,0,1, 1,0,2, 0,1,4, 1,1,8) {
my ($mx, $my) = ($cx+$dx, $cy+$dy);
$mask += $b if $mx>1 and $my>1 and $data[$my-1][$mx-1] != 0
}
$d = N if any { $mask == $_ } (1, 5,13);
$d = E if any { $mask == $_ } (2, 3, 7);
$d = W if any { $mask == $_ } (4,12,14);
$d = S if any { $mask == $_ } (8,10,11);
$d = $p == N ? W : E if $mask == 6;
$d = $p == E ? N : S if $mask == 9;
$path .= $p = (<E N W S>)[$d];
$cx += (1, 0,-1,0)[$d];
$cy += (0,-1, 0,1)[$d];
} until $cx == $x and $cy == $y;
return $x, -$y, $path
}
exit 'That did not work out...';
}
my @M = ([0, 0, 0, 0, 0],
[0, 0, 0, 0, 0],
[0, 0, 1, 1, 0],
[0, 0, 1, 1, 0],
[0, 0, 0, 1, 0],
[0, 0, 0, 0, 0]);
printf "X: %d, Y: %d, Path: %s\n", identify_perimeter(@M);</syntaxhighlight>
{{out}}
<pre>X: 2, Y: -2, Path: SSESENNNWW</pre>
=={{header|Phix}}==
Based on the same code as the Wren example.
<!--<
<span style="color: #008080;">with</span> <span style="color: #008080;">javascript_semantics</span>
<span style="color: #008080;">enum</span> <span style="color: #000000;">E</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">N</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">W</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">S</span>
Line 269 ⟶ 450:
<span style="color: #7060A8;">printf</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"X: %d, Y: %d, Path: %s\n"</span><span style="color: #0000FF;">,</span><span style="color: #000000;">identifyPerimeter</span><span style="color: #0000FF;">(</span><span style="color: #000000;">example</span><span style="color: #0000FF;">))</span>
<!--</
{{out}}
<pre>
Line 276 ⟶ 457:
=={{header|Python}}==
<
from skimage import measure
Line 291 ⟶ 472:
contours = round(measure.find_contours(example, 0.1))[0]
print('[', ', '.join([str((p[1], 5 - p[0])) for p in contours]), ']')
</
<pre>
[ (3.0, 0.0), (2.0, 1.0), (2.0, 1.0), (1.0, 2.0), (1.0, 3.0), (2.0, 4.0), (3.0, 4.0), (4.0, 3.0), (4.0, 2.0), (4.0, 1.0), (3.0, 0.0) ]
</pre>
=={{header|Raku}}==
{{trans|Phix}}
<syntaxhighlight lang="raku" line># 20220708 Raku programming solution
enum <E N W S>;
my (@dx,@dy) := (1,0,-1,0), (0,-1,0,1);
my \example = ((0, 0, 0, 0, 0),
(0, 0, 0, 0, 0),
(0, 0, 1, 1, 0),
(0, 0, 1, 1, 0),
(0, 0, 0, 1, 0),
(0, 0, 0, 0, 0));
printf("X: %d, Y: %d, Path: %s\n", identifyPerimeter(example));
sub identifyPerimeter(\data) {
my (\height,\width) = { .elems, .first.elems }(data);
for ^width X ^height -> (\x,\y) {
if data[y;x] {
my ($cx,$cy,$directions,$previous) = x, y;
repeat {
my $mask;
for (0,0,1),(1,0,2),(0,1,4),(1,1,8) -> (\dx,\dy,\b) {
my ($mx,$my) = $cx+dx,$cy+dy;
$mask += b if all $mx>1, $my>1, data[$my-1;$mx-1]
}
given do given $mask {
when * ∈ ( 1, 5, 13 ) { N }
when * ∈ ( 2, 3, 7 ) { E }
when * ∈ ( 4, 12, 14 ) { W }
when * ∈ ( 8, 10, 11 ) { S }
when * ∈ ( 6 ) { $previous == N ?? W !! E }
when * ∈ ( 9 ) { $previous == E ?? N !! S }
} {
$directions ~= $previous = $_ ;
($cx,$cy) <<+=<< (@dx[.value], @dy[.value])
}
} until $cx==x and $cy==y ;
return x, -y, $directions
}
}
return -1, -1, "Not found!"
}</syntaxhighlight>
Output is the same as the Phix entry.
=={{header|Wren}}==
{{libheader|Wren-seq}}
This is a translation of the [http://www.tomgibara.com/computer-vision/marching-squares public domain Java code], written by Tom Gibara, which is linked to from the Wikipedia article. It also uses his example to test the code.
<
/* A direction in the plane. */
Line 516 ⟶ 744:
var ms = MarchingSquares.new(5, 6, example)
var path = ms.identifyPerimeter()
System.print(path)</
{{out}}
| |||