Filling a region between parametric curves?

4 次查看(过去 30 天)
Hi! I am trying to fill a region between parametric curves, defined by the following code (in a zgrid):
figure; zgrid; hold on;
Ts = .1; wn = [.4*pi/Ts .6*pi/Ts]; FillColor = 'r';
wn_lb = wn(1); wn_ub = wn(2);
% Values of zeta that correspond to start and end of wn curves:
zeta = linspace(0,1);
% Create vector of complex numbers to plot:
mag_lb = exp(-zeta.*wn_lb*Ts);
ang_lb = sqrt(1-zeta.^2).*wn_lb*Ts;
z_lb = mag_lb.*exp(ang_lb*1j);
mag_ub = exp(-zeta.*wn_ub*Ts);
ang_ub = sqrt(1-zeta.^2).*wn_ub*Ts;
z_ub = mag_ub.*exp(ang_ub*1j);
% Create unit circle arc for appropriate shading:
theta_lb = linspace(angle(conj(z_lb(1))),angle(z_lb(1)));
unit_circle_lb = cos(theta_lb) + sin(theta_lb)*1j;
theta_ub = linspace(angle(conj(z_ub(1))),angle(z_ub(1)));
unit_circle_ub = cos(theta_ub) + sin(theta_ub)*1j;
theta = [linspace(angle(conj(z_ub(1))),angle(conj(z_lb(1)))) linspace(angle(z_lb(1)),angle(z_ub(1)))];
unit_circle = cos(theta) + sin(theta)*1j;
% This is a plot of the region I want to get!
scatter([real(z_ub) flip(real(z_ub)) real(unit_circle) real(z_lb) flip(real(z_lb))],...
[imag(z_ub) flip(-imag(z_ub)) imag(unit_circle) imag(z_lb) flip(-imag(z_lb))])
% This is all I've gotten so far... :(
hold off; figure;
fill([real(z_ub),flip(real(z_ub)),(real(unit_circle)),real(z_lb),flip(real(z_lb)),real(unit_circle)], ...
[imag(z_ub),flip(-imag(z_ub)),(imag(unit_circle)),imag(z_lb),flip(-imag(z_lb)),imag(unit_circle)],...
FillColor,'FaceAlpha',.3);
zgrid
However, I can't seem to figure out how to fill the center region of figure 1. Thus far, figure 2 is the best I've gotten to using fill.
(Perhaps using patch?)
Thanks so much!
  1 个评论
Jonathan Bessette
figure; patch([real(z_ub) flip(real(z_ub)) real(unit_circle) real(z_lb) flip(real(z_lb))],...
[imag(z_ub) flip(-imag(z_ub)) imag(unit_circle) imag(z_lb) flip(-imag(z_lb))],'r')
I tried this method, but I can't seem to get the order of the verticies correct.

请先登录,再进行评论。

采纳的回答

Ameer Hamza
Ameer Hamza 2020-4-7
The actual issue is the order of the point. The patch function fails because the points are not distributed as a closed-loop. The following uses a very simple way to order the vertices and then call the patch function.
figure; zgrid; hold on;
Ts = .1; wn = [.4*pi/Ts .6*pi/Ts]; FillColor = 'r';
wn_lb = wn(1); wn_ub = wn(2);
% Values of zeta that correspond to start and end of wn curves:
zeta = linspace(0,1);
% Create vector of complex numbers to plot:
mag_lb = exp(-zeta.*wn_lb*Ts);
ang_lb = sqrt(1-zeta.^2).*wn_lb*Ts;
z_lb = mag_lb.*exp(ang_lb*1j);
mag_ub = exp(-zeta.*wn_ub*Ts);
ang_ub = sqrt(1-zeta.^2).*wn_ub*Ts;
z_ub = mag_ub.*exp(ang_ub*1j);
% Create unit circle arc for appropriate shading:
theta_lb = linspace(angle(conj(z_lb(1))),angle(z_lb(1)));
unit_circle_lb = cos(theta_lb) + sin(theta_lb)*1j;
theta_ub = linspace(angle(conj(z_ub(1))),angle(z_ub(1)));
unit_circle_ub = cos(theta_ub) + sin(theta_ub)*1j;
theta = [linspace(angle(conj(z_ub(1))),angle(conj(z_lb(1)))) linspace(angle(z_lb(1)),angle(z_ub(1)))];
unit_circle = cos(theta) + sin(theta)*1j;
% This is a plot of the region I want to get!
scatter([real(z_ub) flip(real(z_ub)) real(unit_circle) real(z_lb) flip(real(z_lb))],...
[imag(z_ub) flip(-imag(z_ub)) imag(unit_circle) imag(z_lb) flip(-imag(z_lb))])
% ordering the vertices
x = [real(z_ub) flip(real(z_ub)) real(unit_circle) real(z_lb) flip(real(z_lb))];
y = [imag(z_ub) flip(-imag(z_ub)) imag(unit_circle) imag(z_lb) flip(-imag(z_lb))];
X_original = [x' y'];
X_ordered = zeros(size(X_original));
X_ordered(1,:) = X_original(1,:);
x_temp = X_original(1,:);
X_original(1,:) = [];
count = 2;
while ~isempty(X_original)
[~,idx] = min(pdist2(X_original, x_temp));
X_ordered(count,:) = X_original(idx, :);
x_temp = X_original(idx, :);
X_original(idx, :) = [];
count = count + 1;
end
hold off; figure;
p = patch(X_ordered(:,1), X_ordered(:,2), 'r');
p.FaceAlpha = 0.2;
p.EdgeColor = 'none';
zgrid

更多回答(1 个)

darova
darova 2020-4-7
Try this to detect which values in a wrong order
% This is a plot of the region I want to get!
X = [real(z_ub) flip(real(z_ub)) real(unit_circle) real(z_lb) flip(real(z_lb))];
Y = [imag(z_ub) flip(-imag(z_ub)) imag(unit_circle) imag(z_lb) flip(-imag(z_lb))];
for i = 1:10:length(X)-11
plot(X(i:i+10),Y(i:i+10),'linewidth',2)
pause(0.1)
end
  1 个评论
Jonathan Bessette
Thanks for your input! Ameer Hamza gave a detailed explanation, incorporating the idea (which you mentioned) of properly ordering points before using the "patch" function.

请先登录,再进行评论。

类别

Help CenterFile Exchange 中查找有关 Graphics Object Properties 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by