How to get average image of a set of .bmp images?

Hi, all

I am not stuck with a problem. I have a set of .bmp images, which are actually images of faces. I'd like to get the average image of these images. The following is what I do:

clear all
clc

path = 'C:\myImages\';
names = ls(path);
names(1:2,:)=[];

red = zeros(512, 512, 50);
green = zeros(512, 512, 50);
blue = zeros(512, 512, 50);

for i = 1:50
    t = imread([path names(i,:)]);
    red(:,:,1) = double(t(:,:,1));
    green(:,:,2) = double(t(:,:,2));
    blue(:,:,3) = double(t(:,:,3));
end

for i = 1:512
    for j = 1:512
        r(i,j) = mean(red(i,j,:));
        g(i,j) = mean(green(i,j,:));
        b(i,j) = mean(blue(i,j,:));
    end
end

newimage(:,:,1) = uint8(r);
newimage(:,:,2) = uint8(g);
newimage(:,:,3) = uint8(b);

imshow(newimage)

You can see that I try to convert the uint8 pixel into double variables, find the average value of the pixels, then convert the variables back to uint8 type.

However, all I've got is a totally black image. I examine individual pixels of the new image, and the pixel valus are around 0~5.

May I know what is wrong with my code? Many thanks for your help!

Best regards Wenlong

Answer 1

[Image Analyst]

Why not just use the mean function:

meanRedImage = mean(red, 3);
meanGreenImage = mean(green, 3);
meanBlueImage = mean(blue, 3);

meanRGB = uint8(cat(3, meanRedImage, meanGreenImage, meanBlueImage));