How would find the average within each region?

This code generates regions within a user defined image. I want to find the average value within each region and save it as a new image.

clc,clear, close all
% Load the grayscale image
img = imread('peppers.png'); % Replace with your file path
img = rgb2gray(img);
img = mat2gray(img);

threshold = 0.05; % Adjust this value for different levels of similarity
% Ensure the dimensions are a power of 2 by padding
[m, n] = size(img);
M = 2^nextpow2(m); % Nearest power of 2 greater than or equal to m
N = 2^nextpow2(n); % Nearest power of 2 greater than or equal to n
paddedImg = padarray(img, [M-m, N-n], 'post');

% Initialize the quadtree decomposition
quadtreeDecomp(paddedImg, threshold);

function quadtreeDecomp(img, threshold)
    % Display the image
    imshow(img, 'InitialMagnification', 'fit');
    hold on;
    % Call recursive function for quadtree decomposition
    recursiveSplit(img, [1, 1], size(img, 1), threshold);
    hold off;
end

function recursiveSplit(img, origin, sizeBlock, threshold)
    % Extract the current block
    x = origin(1);
    y = origin(2);
    block = img(x:x+sizeBlock-1, y:y+sizeBlock-1);

    % Check homogeneity (using variance)
    if std2(block) > threshold && sizeBlock > 4
        % Split into four sub-blocks
        halfSize = sizeBlock / 2;

        % Recursively process each sub-block
        recursiveSplit(img, [x, y], halfSize, threshold);             % Top-left
        recursiveSplit(img, [x, y+halfSize], halfSize, threshold);    % Top-right
        recursiveSplit(img, [x+halfSize, y], halfSize, threshold);    % Bottom-left
        recursiveSplit(img, [x+halfSize, y+halfSize], halfSize, threshold); % Bottom-right
    else
        % Draw rectangle for this block
        rectangle('Position', [y, x, sizeBlock, sizeBlock], 'EdgeColor', 'r', 'LineWidth', 0.5);
    end
end

Answer 1

[Image Analyst]

Try (untested)

function blockMean = recursiveSplit(img, origin, sizeBlock, threshold)
% Extract the current block
x = origin(1);
y = origin(2);
block = img(x:x+sizeBlock-1, y:y+sizeBlock-1);
% Check homogeneity (using variance)
if std2(block) > threshold && sizeBlock > 4
    % Split into four sub-blocks
    halfSize = sizeBlock / 2;
    % Recursively process each sub-block
    recursiveSplit(img, [x, y], halfSize, threshold);             % Top-left
    recursiveSplit(img, [x, y+halfSize], halfSize, threshold);    % Top-right
    recursiveSplit(img, [x+halfSize, y], halfSize, threshold);    % Bottom-left
    recursiveSplit(img, [x+halfSize, y+halfSize], halfSize, threshold); % Bottom-right
    blockMean = []; % Don't return a mean if we're still splitting up the block.
else
    % Draw rectangle for this block
    rectangle('Position', [y, x, sizeBlock, sizeBlock], 'EdgeColor', 'r', 'LineWidth', 0.5);
    blockMean = mean(block, 'all'); % No splitting needed, so return the mean.
end
end

[Image Analyst]

From the documentation for qtdecomp:

% Read image into the workspace.
I = imread('liftingbody.png');
% Perform the quadtree decomposition and display the block representation in a figure.
S = qtdecomp(I,.27);
blocks = repmat(uint8(0),size(S));

for dim = [512 256 128 64 32 16 8 4 2 1];    
    numblocks = length(find(S==dim));    
    if (numblocks > 0)        
        values = repmat(uint8(1),[dim dim numblocks]);
        values(2:dim,2:dim,:) = 0;
        blocks = qtsetblk(blocks,S,dim,values);
    end
end

blocks(end,1:end) = 1;
blocks(1:end,end) = 1;

imshow(I)

[James hart]

Thank you so much. Your answer worked perfectly.