Creating Binary Plot From Data

I am trying to use conditional statements to create a plot from a binary matrix. I cannot get matlab to create a binary matrix; let alone create the plot. The code I have thus far:

format long
dist;
a = [new_data{1, 1}.subj_xPath.',new_data{1, 1}.subj_yPath.'];
b = [new_data{1, 1}.eye_X.',new_data{1, 1}.eye_Y.'];
dist = sqrt((a(:,1) - b(:,1)).^2 + (a(:,2) - b(:,2)).^2);

nrows = 133;
ncols = 1;
A = ones(nrows,ncols);

for p = 1:ncols
    for q = 1:nrows

          if dist <= 100
              A(q,p) = 1;
          else
              A(q,p) = 0;
          end

      end
  end

I hope to create an plot that looks similar to

Thank you in advance!

Answer 1

[Star Strider]

I’m not certain what you want to do, and I can’t run your code.

For the plot, see if the spy function will do what you want.

[Courtney Thomas, Jr.]

I updated the code for you to try running again.

1. From a data sets 'a' and 'b', which are xy coordinates in a 133x2, I calculated distance. I am now trying to use conditional statements to tell matlab "Create a binary matrix. For each distance ≤ 100 input a 1 and if greater a 0.

2. Once a binary matrix is created I would like to plot it to look like the figure above. If a 1, a bar is plotted; if 0, it is left blank.

[Star Strider]

The problem is here:

if dist <= 100

Your ‘dist’ variable is a column vector, and at each step you are testing the entire vector.

What do you want to do?

[Courtney Thomas, Jr.]

I would like for matlab to give me a 1 or 0 for each item in the "dist" vector. For each distance ≤ 100 input a 1 and if greater a 0.

[Star Strider]

That I understand.

I have a Neural Network Toolbox called dist, so changing it to ‘dst_v’, you can use the hypot function:

dst_v = hypot((a(:,1)-b(:,2)), (a(:,2)-b(:,2)));

How do you want to index it? As you have written your code, ‘A’ is going to be a vector that replicates ‘dist’:

A = dst_v <= 100;

That will produce a logical vector. To produce a numeric vector:

A = +(dst_v <= 100);

You can use reshape to create a matrix from it, but I don’t know what dimensions you want your matrix to be.

Answer 2

[Image Analyst]

I'm not sure why your desired plot is a 1-D plot but you're creating a 2D array for "A". Why is that? And your desired plot is a tri or quad level plot, not a binary one. Why is that? And what are A and dist? Are they 1-D vectors or 2-D matrices? About all I can figure from your code is this:

format long
dist;
a = [new_data{1, 1}.subj_xPath.',new_data{1, 1}.subj_yPath.'];
b = [new_data{1, 1}.eye_X.',new_data{1, 1}.eye_Y.'];
dist = sqrt((a(:,1) - b(:,1)).^2 + (a(:,2) - b(:,2)).^2);
A = dist <= 100; % Binary or matrix
if ndims(A) == 1
    % A is a vector
    plot(A, 'b-');
    grid on;
elseif ndims(A) == 2
    % A is a 2-D array
    imshow(A, []);
    axis on;
end

[Courtney Thomas, Jr.]

A should be a 1-D array. Thanks for pointing that out. My desired plot is a tri or quad level plot because I need to create binary matrices for more than one data sets. I hope to plot each on one graph.

A is the binary matrix produced. dist are the 1x1 calculated results from the

sqrt((a(:,1) - b(:,1)).^2 + (a(:,2) - b(:,2)).^2)
%The distance between point in 'a' and point in 'b'.

[Image Analyst]

What is new_data? We can't run your code because you didn't supply it. Is it a table, a cell array, a structure, structures inside cells of a cell array? Whatever, it's kind of complicated to figure out so it's best if you just supply it to us.

With your code, which is equivalent to my code

A = dist <= 100;

You're going to end up with a binary vector with 2 levels, not a tri- or quad-leveled matrix. Why do you think it should? What ranges would you want to use to set the output to the 3 or 4 output levels, like 0-1 gets mapped to 1, 1-4 gets mapped to 2, 4-20 gets mapped to 3, or whatever. You have to say what ranges get mapped to each of the 3 or 4 levels you want.