what does the four vector mean in matlab legend 'Position'?

75 次查看（过去 30 天）
Martin
Martin 2013-7-3
评论： Malek Barhoush 2020-10-24
I use the line:
leg = legend('Ionisation','Recombination');
pos = get(leg,'Position')
But get returned the answer:
pos =0.638988095238095,0.209126984126984, O.257142857142857,0.1
Can I ask what this means because this not the positions on the axes because when you look by eye it seems incorrect...
All I want to do is find the location (on the axes of the graph) of the bottom left corner so I can add another text box below with a string in it...
Thanks in advance MA

请先登录，再回答此问题。

回答（4 个）

Azzi Abdelmalek
Azzi Abdelmalek 2013-7-3
pos(1) and pos(2) are the coordinate of the legend windows
pos(3) and pos(4) are width and height of this windows
  3 个评论
显示 1更早的评论
Azzi Abdelmalek
Azzi Abdelmalek 2013-7-4
To understand try:
x0=0;
y0=0;
width=0.25;
height=0.25;
plot(1,1,2,2);
legend({'a','b'},'position',[x0 y0 width height])
change the position x0, and y0 then run again
x0=0.5;
y0=1
legend({'a','b'},'position',[x0 y0 width height])
x0,y0 are the coordinate of the left down corner of the legend windows
Malek Barhoush
Malek Barhoush 2020-10-24
x0=0;
y0=0;
width=0.25;
height=0.25;
plot(1,1,2,2);
legend({'a','b'},'position',[x0 y0 width height])

请先登录，再进行评论。

Walter Roberson
Walter Roberson 2013-7-4
In order to understand the coordinates, you need to get() the Units property. Units can be any of several things, but in my experience the three that are used the most are "normal", "pixel", and "data".
Pixel is a measurement in pixels; fractional pixels are allowed.
Normal is based on fractions of the containing object, 0 being none of the object and 1 being the entire object. For example 'normal' and 0.3 as an axis position would mean 3/10 of the way along (from the left) the containing figure (or uipanel)
Data is data coordinates, based around the current xlim, ylim, or zlim. For example in your call
plot(1,1,2,2)
the 1's and 2's are in data coordinates.
If a graphics object has no Units property but has a Position property, then the graphics object is locked into Data coordinates; those kinds of graphics objects must be children of an axes.
Jan
Jan 2013-7-4
编辑：Jan 2013-7-4
You can learn more about the position by experimenting:
leg = legend('Ionisation','Recombination');
pos = get(leg,'Position');
pause(1);
set(leg, 'Position', [0.64, 0.21, O.25, 0.1]);
pause(1);
set(leg, 'Position', [0.74, 0.21, O.25, 0.1]);
pause(1);
set(leg, 'Position', [0.64, 0.31, O.25, 0.1]);
pause(1);
set(leg, 'Position', [0.64, 0.21, O.35, 0.1]);
pause(1);
set(leg, 'Position', [0.64, 0.21, O.25, 0.2]);
pause(1);
set(leg, 'Position', [0.0, 0.0, 1.0, 1.0]);
pause(1);
set(leg, 'Position', [0.5, 0.5, 0.4, 0.4]);
Martin
Martin 2013-7-4
Thanks for all the information on this guys. I think what I have is the dimensions given in 'normal' units. Basically I have to run this simulation a lot of times, each time the y axis can change by 2 orders of magnitude and therefore I can't hardwire the textbox positions. I'm hoping that I can find some sort of conversion factor between 'normal' and 'data' coords as that's what the textbox positions are given in.
  2 个评论
Jan
Jan 2013-7-4
编辑：Jan 2013-7-4
The limits of the axes do not matter, whan you use 'normalized' posiotions. Matlab converts the absolute position on the screen automatically for you. This code writes the string to the upper right corner without knowing the data size:
axes('XLim', cumsum(rand(1, 2)), 'YLim', cumsum(rand(1, 2)));
H = text(0.9, 0.9, 'Hello', 'VerticalAlignment', 'top', ...
'HorizontalAlignment', 'right', ...
'Units', 'normalized');
drawnow;
% Obtain the position in data units if wanted:
set(H, 'Units', 'data');
dataPos = get(H, 'Position');
Martin
Martin 2013-7-4
Ah that's a really good way to do it, thanks. I appreciate that a lot, and certainly have learned a good deal.

请先登录，再进行评论。

类别

在 Help Center 和 File Exchange 中查找有关 Legend 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