how to set a marker at one specific point on a plot (look at the picture)?

Answer 1

[Mischa Kim]

Benjamin, add

 hold on
 plot(x_pos,y_pos,'r*')

to your code. (x_pos,y_pos) is the data point that needs to be marked. So if you have

 x = 0:0.1:pi;
 y = sin(x);

you could use

 plot(x,y)
 hold on
 plot(x(10),y(10),'r*')  % marking the 10th data point of x and y

Starting in R2019a, you can pin multiple data tips. Data tips appear temporarily as you hover over the data points in a figure window. To display persistent(pinned) data tips, click one or more data points.

For more information on "Interactively explore Plotted Data", please refer to the following documentation:

https://www.mathworks.com/help/releases/R2019a/matlab/creating_plots/interactively-explore-plotted-data.html

[benjamin ma]

Thank you. it works now. but one question, can i use "set" function to mark a plot also?

[Matt Kindig]

Sure. First you will need to define the plot object, however. Something like this should work:

h = plot(NaN, NaN, 'r*');  %define the graphics object, but don't render anything.
set(h,'Xdata', x(10), 'YData', y(10));   %mark 10th point
pause(5);   %after a few seconds...
set(h, 'XData', x(20), 'YData', y(20));  %mark 20th point instead.

[Vijay shankar Sridharan]

Hi Mischa, I get error when I try to plot points with decimal points.

[Walter Roberson]

Could you expand on what you mean about plotting points with decimal points? For example are you trying to do

set(h, 'XData', x(20.3), 'YData', y(20.3));  %mark 20.3th point instead.

Answer 2

[Greg]

With 7k views in a month, I'm surprised this hasn't been updated.

Starting in R2016b, there is a MarkerIndices property. Instead of the other answer's suggested:

 plot(x,y);
 hold on;
 plot(x(10),y(10),'r*');

Now simply use:

h = plot(x,y,'MarkerIndices',10);

Move the marker(s) around at any time if you've stored the handle h:

h.MarkerIndices = 5:5:length(x);

[MathWorks Support Team]

In addition, you can specify the ‘o-‘ line style, which creates a solid line and markers. You can also specify marker properties, such as the face color and edge color.

x = linspace(0,pi,30);
y = sin(x);
p = plot(x,y,'o-','MarkerFaceColor','red','MarkerEdgeColor','red','MarkerIndices',10)

Answer 3

[navi g]

hello, is this marking in plot is possible without writing code, and putting mark in plot in figure editor,

and for sinosoidal curve i have only x data, i dont have y data, but i need to mark on curve, means that i will give x value, based on x value it should place on curve exactly on sinosoidal curve,

[Walter Roberson]

For R2014b or later (I would have to check about earlier; I see some references in 2012 time frame) you can use data brushing. Click on the paintbrush in the figure and then you can click on a point to mark it.

For marking a particular location given only the x, then

x_to_mark = SomeSpecificXValue;

all_lines = findobj(gca, 'type', 'line');
number_of_marks = 0;
where_to_mark = [];
for K = 1 : length(all_lines)
  this_line = all_lines(K);
  this_xdata = get(this_line, 'XData');
  if x_to_mark < min(this_xdata) | x_to_mark > max(this_xdata)
     continue;   %the line does not span that x
  end
  x_diff = diff(this_xdata);
  if isempty(x_diff)
     fprintf('skipping line #%d that is single point\n', K);
  elseif all(x_diff > 0) | all(x_diff < 0)
     %it is monotonic, safe to do interpolation
     this_ydata = get(this_line, 'YData');
     y_to_mark = interp1(this_xdata, this_ydata, x_to_mark);
     number_of_marks = number_of_marks + 1;
     where_to_mark(number_of_marks,:) = [x_to_mark, y_to_mark];
  else
     fprintf('skipping line #%d with unsorted x data\n', K);
  end
end

if number_of_marks == 0
  fprintf('That x was not found on a line we could handle\n');
else
  hold on
  plot(where_to_mark(:,1), where_to_mark(:,2), 'r*');
end

This is probably a lot longer than you were expecting. You did not happen to provide information that we might potentially have used to make it shorter. For example if you know there is only exactly one line, and that the x were sorted when you plotted, and that the x value to mark is definitely in range, then the code could be made much shorter.