Extracting data from Contour plots

Question

Mohammed Darwish 2017-5-18

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/340853-extracting-data-from-contour-plots

评论： Star Strider 2022-7-2

I am trying to extract data from a contour plot. I managed to get all the information about the dataObjs using

 open('colormapinhomo.fig');
 h = gcf;
 axesObjs = get(h, 'Children');
 dataObjs = get(axesObjs, 'Children')
 dataObjs{2}
Contour with properties:
    LineColor: [0 0 0]
    LineStyle: 'none'
    LineWidth: 0.5000
         Fill: 'on'
    LevelList: [1×62 double]
        XData: [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40]
        YData: [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20]
        ZData: [20×40 double]

but cant get the data itself. Any help ? And is this the only way to extract data, or can I somehow output a .dat file of the figure with the data whether it was a contour or a histogram or even a simple plot line ?

2 个评论
显示无隐藏无

Adam 2017-5-18

在 MATLAB Online 中打开

dataObjs{2}.XData
dataObjs{2}.YData
dataObjs{2}.ZData

is the data. Generally you should keep hold of the data when you plot because extracting data from a plot is silly when you can just use the original data. If you've lost the original data though then the above is the best you can get from the plot.

Adam Danz 2020-1-23

Also see this file exchange function that extracts the coordinates of all contour lines and organizes them in a table by level.

https://www.mathworks.com/matlabcentral/fileexchange/74010-getcontourlinecoordinates

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Star Strider 2017-5-18

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/340853-extracting-data-from-contour-plots#answer_267508

在 MATLAB Online 中打开

If you want the contour (x,y) values, you have to ask for them:

figure(1)
[C,h] = contour( ... );

The (2xN) ‘C’ array will have the (x,y) values of the contour lines. The first column in each has the contour value as the first row and the number of (x,y) values defining the contour as the second row. (It requires some coding to extract all of them effectively.) Use the LevelList property to find the values of the contours (these will match the first row and first column of each contour) so you can then use find to locate the beginning of each (x,y) contour matrix.

See the documentation for contour for details. There is a link to ‘Contour Properties’ that lists all of them.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Star Strider 2022-7-2

在 MATLAB Online 中打开

@Fereshteh —

With respect to your Comment, the centroid is the mean of the ‘x’, ‘y’, (and ‘z’) coordinates, so just take the mean values of the vectors —

[X,Y,Z] = peaks;

figure

[C,h] = contour(X,Y,Z, [1 1]*2); % Find One Contour At Level = 2

Cix = find(C(1,:)==2);

for k = 1:numel(Cix)

ixr = Cix(k)+1 : Cix(k)+C(2,Cix(k));

Cx{k} = C(1,ixr);

Cy{k} = C(2,ixr);

Crtd(:,k) = [mean(Cx{k}); mean(Cy{k})];

end

Crtd % Centroid Calculated From ‘mean’ Values

Crtd = 2×2

-0.5006 0.5180 -0.5978 1.0134

hold on

for k = 1:numel(Cix)

% plot(Cx{k}, Cy{k})

plot(Crtd(1,k), Crtd(2,k), 'rx')

end

hold off

title('Centroid Calculated From ‘mean’ Values')

figure

hold on

for k = 1:numel(Cix)

ps = polyshape(Cx{k}, Cy{k});

[psCtrdx(:,k),psCtrdy(:,k)] = centroid(ps);

plot(ps)

plot(psCtrdx,psCtrdy, 'rx');

end

psCrtd = [psCtrdx; psCtrdy] % Centroid Calculated By The ‘centroid’ Function

psCrtd = 2×2

-0.4788 0.4495 -0.5980 1.1698

hold off

axis([-3 3 -3 3])

title('Centroid Calculated By The ‘centroid’ Function')

The mean values and the centroid values are approximately the same.

This is a bit more involved for more contours (and requires a loop for each contour nesting each of these for that reason) however the approach is the same.

.

请先登录，再进行评论。