how to divide an image into 4*4 blocks and find the mean of each block ?

we are doing our graduation project on depth map generation.we have read an coloured image and converted it into a grsay scale.next we have to divide the image into 4*4 blocks and find the mean of each block.

Answer 1

[Jan]

Img = rand(640, 480);  % Test data
Img = reshape(Img, 4, 160, 4, 120);
meanImg = sum(sum(Img, 1), 3) / 16;
meanImg = squeeze(meanImg)

[maria]

meanImg = sum(sum(Img, 1), 3) / 16; wat does this statement do??

[maria]

thanks a lot.. can u pls tell me how to find the mean of each 4*4 block after the following codes

clc;clear all;close all;

i=importdata('C:\Users\Public\Pictures\Sample Pictures\Koala.jpg'); Img=rgb2gray(i);

 [r c]=size(Img);
x=r/4;y=c/4;
 Img = reshape(Img, 4, x, 4, y);
meanImg = sum(sum(Img, 1), 3) / 16;
meanImg = squeeze(meanImg);

[Jan]

@Maria: This statement calculates the sum twice, at first along the 1st dimension, than along the 3rd dimension. You can test this step by step using a rand(8,8) matrix.

I do not understand, what you want to get. What does "group" exactly mean here?

[maria]

how to find the differnce between all the adjacent elements in an array???

Answer 2

[Matt J]

Using this

http://www.mathworks.com/matlabcentral/fileexchange/35085-split-nd-array-into-equal-sized-cells

 meanImg = cellfun(@(c) mean(c(:)),   mat2tiles(rgb2gray(Img),[4,4]) );

[maria]

i am not able to acess MAT2TILES fn and wat is 'c'?

[maria]

THANKS

Answer 3

[Thorsten]

imshow(blkproc(rgb2gray(I), round([size(I,1) size(I, 2)]/4), ...
       @(x) mean2(x)*ones(size(x)) ))