Convex Hull algorithm without built-in function
2 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
I have found on mathworks a convex hull algorithm, which uses randomly defined vectors as an input. I have tried to change random vectors to another ones, which I want to use. Unfortunately, the program says 'indeces on the right sight are not compatible with indeces on left side', giving me error at ind(1) = find(V == P1(:,1));
And I don't know what I need to change to make this program work with user-defined vectors, not random ones. So, I would be very glad for any help with this question.
x = random('Normal',0,1,1,10);
y = random('Normal',0,1,1,10);
InputData = [x',y'];
[V,I,vertices] = ConvexHull(InputData);
figure(46)
plot(InputData(:,1),InputData(:,2),'.','Color','red');
hold on
plot(V(:,1),V(:,2),'Marker','^','LineStyle','--','Color','#7E2F8E')
title('SuperOutput');
xlabel('x');
ylabel('y');
legend('InputData','Convex hull');
function [V,I,n] = ConvexHull(InputData)
V = [InputData(1:3,:)]; I = [1:3]';
for k = 4:length(InputData)
pk = InputData(k,:);
sup_ver = [];
ind = [];
index = [];
for i = 1:length(V)
vi = V(i,:);O = [];
for j = 1:length(V)
if (i ~= j)
vj = V(j,:);
O(j) = det([1,pk;1,vi;1,vj]);
end
end
O(O == 0) = [];
if ((numel(unique(sign(O))) == 1) == 1)
sup_ver = [sup_ver;vi];
end
end
if (isempty(sup_ver) == 0)
P1 = sup_ver(1,:);
P2 = sup_ver(2,:);
P3 = pk;
s = det([P1 - P2;P3 - P1]);
n = 1;
while (n <= length(V))
P = V(n,:);
if ((P ~= P1) & (P ~= P2) & (P ~= pk) & ((s * det([P3 - P;P2 - P3]) >= 0) & (s * det([P1 - P;P3 - P1]) >= 0) & (s * det([P2 - P;P1 - P2]) >= 0)) == 1)
V(n,:) = [];
I(n) = [];
n = n - 1;
end
n = n + 1;
end
ind(1) = find(V == P1(:,1));
ind(2) = find(V == P2(:,1));
index = find(InputData == pk(:,1));
if (ind(2) - ind(1) == 1)
V = [V(1:ind(1),:);pk;V(ind(1) + 1:end,:)];
I = [I(1:ind(1));index;I(ind(1) + 1:end)];
else
V = [pk;V];
I = [index;I];
end
end
end
V = vertcat(V,V(1,:));
I = vertcat(I,I(1));
end
0 commentaires
Réponses (0)
Voir également
Catégories
En savoir plus sur Bounding Regions dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!