How to make co-occurence Matrix Like this (as shown in Image)

I have data points for one parameter A (200 to 800) at every 5 unit resolution and another paramemter B (40 to 200) at every 1 km resolution. I would like to know how to make normalized Co occurence matrix like in the image (attached).

Answer 1

[Sourabh]

Hey @Arun Kumar Singh

I will help you on making a dummy normalised co-occurrence matrix as per the image by breaking it down into 2 steps – Computing the co-occurrence matrix followed by visualizing it as a heatmap.

1. Computing the co-occurrence matrix

• Define bins for parameters A and B:

A_bins = 200:5:800; % A ranges from 200 to 800 at a 5-unit resolution 
B_bins = 40:1:200; % B ranges from 40 to 200 at a 1-km resolution 

• Loading data: If you have real data, load it from a file (csv, mat, etc.). Assuming you have loaded the data for the 2 parameters in A_data and B_data
• Make Co-occurance matrix using “histcounts2” function

occurrence_matrix = histcounts2(A_data, B_data, A_bins, B_bins); 

• Normalize to range [0, 1]

normalized_matrix = occurrence_matrix / max(occurrence_matrix(:)); 

2. Visualizing the co-occurrence matrix as a heatmap

MATLAB provides multiple functions to plot heatmaps. Kindly follow any one the method mentioned below:

• Using “imagesc”

figure; 
imagesc(A_bins, B_bins, normalized_matrix');  
colormap(jet);  
colorbar; 
caxis([0 1]);

• Using “pcolor”

figure; 
pcolor(A_bins(1:end-1), B_bins(1:end-1), normalized_matrix'); 
colormap(jet);  
colorbar; 
caxis([0 1]);  

• Using “contourf”

figure; 
contourf(A_bins(1:end-1), B_bins(1:end-1), normalized_matrix', 20); % 20 contour levels 
colormap(jet);   
colorbar;         
caxis([0 1]); 

Then proceed with labelling the axes accordingly.

For more information on “histcounts2”, Kindly follow the MATLAB documentation:

https://www.mathworks.com/help/matlab/ref/histcounts2.html

For more information on displaying heatmaps, kindly refer the following MATLAB documentations:

https://www.mathworks.com/help/matlab/ref/imagesc.html

https://www.mathworks.com/help/matlab/ref/pcolor.html

https://www.mathworks.com/help/matlab/ref/contourf.html