DSPRelated.com
Forums

Brightness value

Started by cole...@gmail.com August 5, 2009
I'm working with about 1000 images, which are read into matlab as matrix. But due to variation in the intensity of light(weather condition, AC light) during the time the pictures are taken, there is a variation in the pixel value reported in the images. I want to make all pictures to have the same brightness. This means if I should index the image matrix at a point, all the images should have the same value. How can do it? I don't have image processing toolbox, so i depend on writing the code directly into Matlab.

look forward to your reply
I've carried out mean of each image and i've determined the highest mean
from the images. See my code as stated below. However, i've a concern on how
to determine which of the images gave the highest mean reported. Can anyone
advise me on how to do this. code's below
jpegFiles = dir('*.jpg'); % directory of pictures
for k = 1:length(jpegFiles) % loop to run large number of pictures
filename = jpegFiles(k).name;
data1 = imread(filename);
reddata = data1(:,:,1);
z(k+1) = mean(mean(reddata));
a = max(z) % reports the highest mean
end

look forward to your reply

On Thu, Aug 6, 2009 at 3:58 AM, yrodzii wrote:

> --- In m..., colenajayi@... wrote:
> >
> > I'm working with about 1000 images, which are read into matlab as matrix.
> But due to variation in the intensity of light(weather condition, AC light)
> during the time the pictures are taken, there is a variation in the pixel
> value reported in the images. I want to make all pictures to have the same
> brightness. This means if I should index the image matrix at a point, all
> the images should have the same value. How can do it? I don't have image
> processing toolbox, so i depend on writing the code directly into Matlab.
> >
> > look forward to your reply
> >
> One idea is to use the average of all pixels (i.e. mean(your_matrix,3),
> where your matrix is MxNx1000) if this is possible.
>
> On the other hand this fluctuation is normal and might be useful for your
> analysis later.. Just a though!
>
> Another idea is to select the value you want to use and set it for all
> matrices (your_matrix(m,n,:)=n0).
>
> Hope this works...