How can I create a surface plot from ROIs (9) in an image?

I need to input an image, create 9 equal ROIs, and output a surface plot of pixel x pixel to show the 9 ROIs

[Umar]

Hi Gabriel,

Start by loading your image into MATLAB using the `imread` function. Make sure the image is in a format that MATLAB can read, such as JPEG or PNG. Then, divide the image into 9 equal ROIs, you can use the `imdivide` function in MATLAB. This function allows you to specify the number of rows and columns for the division. For example, if your image is a 300x300 matrix, you can divide it into 3 rows and 3 columns to get 9 equal ROIs. Once you have created the 9 ROIs, you can generate a surface plot of pixel x pixel to visualize them. Use the `surf` function in MATLAB to plot the pixels of each ROI on a 3D surface. You can customize the appearance of the plot by adjusting parameters such as color mapping and shading.

By following these steps, you will be able to input an image in MATLAB, create 9 equal ROIs, and output a surface plot to display them effectively.

Hope that answers your question.

[Gabriel]

I forgot to add the 9 ROIs have to be at exact positions in my image, and have to be equivalent. Still no sure how to write out "ROI" in the script besides "ROI = ". I need 9 100 x 100 rectangles at specific regions of interest (in pixels). My image is 1000 x 800 (pixels), and has different intensities in the 9 regions I want to highlight, and then once highlighted, I need this outlined in a surface plot that shows the varying surfaces of the image at each of the 9 ROIs.

[DGM]

To be clear, you want to display the image content selected by each mask region, right?

Does it make more sense to display all the subimages together in a single surf?

% input
inpict = imread('cameraman.tif');

% parameters
roisize = [100 100]; % [y x]
roioffset = [10 10; 128 128]; % [y x]

% generate a mask somehow
nblobs = size(roioffset,1);
mask = false(size(inpict));
for k = 1:nblobs
    rows = roioffset(k,1):roioffset(k,1) + roisize(1) - 1;
    cols = roioffset(k,2):roioffset(k,2) + roisize(2) - 1;
    mask(rows,cols) = true;
end

% apply the mask
Z = inpict;
Z(~mask) = 0; % for float and unsigned-integer images

% show it as an image so we know what we're looking at
figure(1)
imshow(Z)

% display it as a surf
figure(2)
surf(Z)
shading flat

Or would it be easier to read if the subimages are displayed individually?

% input
inpict = imread('cameraman.tif');

% parameters
roisize = [100 100]; % [y x]
roioffset = [10 10; 128 128]; % [y x]

% directly get image content in each blob 
% instead of bothering with a mask
nblobs = size(roioffset,1);
subimages = cell(nblobs,1);
for k = 1:nblobs
    rows = roioffset(k,1):roioffset(k,1) + roisize(1) - 1;
    cols = roioffset(k,2):roioffset(k,2) + roisize(2) - 1;
    subimages{k} = inpict(rows,cols);
end

% display each subimage as a surf
for k = 1:nblobs
    % you might want to do a more thoughtful figure setup
    figure
    
    % the surf
    surf(subimages{k})
    shading flat
end

[Gabriel]

Your method of displaying all the subimages together in a single surf worked perfectly, thank you very much for your help!

Answer 1

[Umar]

Hi Gabriel,

To add 9 ROIs with specific positions in Matlab, you can define the positions of the rectangles using the rectangle function.

For more information on rectangle function, please refer to

https://www.mathworks.com/help/matlab/ref/rectangle.html#

Here's a sample code snippet to create and outline the ROIs on your image:

% Define the positions of the 9 ROIs (x, y, width, height)

roiPositions = [x1, y1, 100, 100; x2, y2, 100, 100; ...]; % Define all 9 ROIs

% Display the image

imshow(yourImage);

hold on; % To overlay rectangles on the image

% Loop through each ROI position and draw the rectangle

for i = 1:size(roiPositions, 1) rectangle('Position', roiPositions(i, :), 'EdgeColor', 'r', 'LineWidth', 2); end

hold off; % Release the hold

% Create a surface plot for each ROI

for i = 1:size(roiPositions, 1) roi = yourImage(roiPositions(i, 2):roiPositions(i, 2)+99, roiPositions(i, 1):roiPositions(i, 1)+99); figure; surf(roi); title(['Surface Plot for ROI ', num2str(i)]); end

The above snippet code will help you add 9 ROIs as rectangles at specific positions in your image and generate surface plots for each ROI to visualize the varying intensities.