Prev: How do you do a moving median fast?
Next: How to align dates in Matlab fast?
From: Luna Moon on 5 Jul 2010 12:20
Hi all,

In recursive moving or rolling average,

result_t = alpha * p_t + (1-alpha)* result_{t-1},

where p_t is the new arrival...

How do we do this fast in Matlab, using the "filter" function format,
i.e. it works on a whole array column-wise.

Now, how to apply the same concept to rolling variance?

I guess in order to do rolling variance, we need to first get the
rolling average,

and then the rolling mean square:

rolling_mean_square_t = alpha*p_t^2 + (1-
alpha)*rolling_mean_square_{t-1},

then using:

rolling_variance_t = rolling_mean_square_t - rolling_mean_t ^2 ...

Am I right? How to do this fast?

Thanks
From: ImageAnalyst on 5 Jul 2010 12:33
Study my variance filter demo and write back if you have any questions
after that:

% Demo to take the local mean, variance, and standard deviation
% of a gray scale image.
% userImage, if passed in, is used as the image.
% If userImage is not passed in, user is asked to use a demo image.
% Code written by ImageAnalyst
function local_variance(userImage)
% Clean up.
clc; % Clear the command window.
close all; % Close all figures (except those of imtool.)
workspace; % Make sure the workspace panel is showing.

% Change the current folder to the folder of this m-file.
% (The line of code below is from Brett Shoelson of The Mathworks.)
% Don't use these lines if you're calling this from another m-file.
if(~isdeployed)
cd(fileparts(which(mfilename)));
end

% Initialize.
fontSize = 20;
if nargin == 0
% No image passed in on the command line.
% Read in one of the standard MATLAB demo images
% as our original gray scale image and display it.
promptMessage = sprintf('Which image do you want to use.\nThe
coins or the cameraman?');
button = questdlg(promptMessage, 'Select Image', 'Coins',
'Cameraman', 'Coins');
if strcmp(button, 'Coins')
grayImage = double(imread('coins.png')); % Cast to double.
else
grayImage = double(imread('cameraman.tif')); % Cast to double.
end
else
% Use the image array passed in on the command line.
grayImage = double(userImage); % Cast to double.
end

% Start timing.
startTime = tic;

subplot(2, 2, 1);
imshow(grayImage, []);
title('Original Image', 'FontSize', fontSize);
set(gcf, 'Position', get(0,'Screensize')); % Maximize figure.

% Blur the image with a 5 by 5 averaging (box filter) window.
blurredImage = conv2(grayImage, ones(5,5)/25);
subplot(2, 2, 2);
imshow(blurredImage, []);
title('Blurred Image', 'FontSize', fontSize);

% Perform a variance filter.
% Output image is the variance of the input image in a 3 by 3 sliding
window.
VarianceFilterFunction = @(x) var(x(:));
varianceImage = nlfilter(grayImage, [3 3], VarianceFilterFunction);
% An alternate way of doing the variance filter is on the next line:
% varianceImage = reshape(std(im2col(originalImage,[3 3],'sliding')),
size(originalImage)-2);
subplot(2, 2, 3);
imshow(varianceImage, [])
title('Variance Image', 'FontSize', fontSize);

% Compute the square root of the variance image to get the standard
deviation.
standardDeviationImage = sqrt(varianceImage);
subplot(2, 2, 4);
imshow(standardDeviationImage, [])
title('Standard Deviation Image', 'FontSize', fontSize);
elapsedTime = toc(startTime);

message = sprintf('Done!\n\nElapsed time = %.2f seconds.',
elapsedTime);
msgbox(message);
return; % End of local_variance() function.
From: Luna Moon on 5 Jul 2010 12:37
On Jul 5, 12:33 pm, ImageAnalyst <imageanal...(a)mailinator.com> wrote:
> Study my variance filter demo and write back if you have any questions
> after that:
>
> % Demo to take the local mean, variance, and standard deviation
> % of a gray scale image.
> % userImage, if passed in, is used as the image.
> % If userImage is not passed in, user is asked to use a demo image.
> % Code written by ImageAnalyst
> function local_variance(userImage)
> % Clean up.
> clc;    % Clear the command window.
> close all;  % Close all figures (except those of imtool.)
> workspace;  % Make sure the workspace panel is showing.
>
> % Change the current folder to the folder of this m-file.
> % (The line of code below is from Brett Shoelson of The Mathworks.)
> % Don't use these lines if you're calling this from another m-file.
> if(~isdeployed)
>         cd(fileparts(which(mfilename)));
> end
>
> % Initialize.
> fontSize = 20;
> if nargin == 0
>     % No image passed in on the command line.
>     % Read in one of the standard MATLAB demo images
>     % as our original gray scale image and display it.
>     promptMessage = sprintf('Which image do you want to use.\nThe
> coins or the cameraman?');
>     button = questdlg(promptMessage, 'Select Image', 'Coins',
> 'Cameraman', 'Coins');
>     if strcmp(button, 'Coins')
>         grayImage = double(imread('coins.png')); % Cast to double.
>     else
>         grayImage = double(imread('cameraman.tif')); % Cast to double.
>     end
> else
>     % Use the image array passed in on the command line.
>     grayImage = double(userImage); % Cast to double.
> end
>
> % Start timing.
> startTime = tic;
>
> subplot(2, 2, 1);
> imshow(grayImage, []);
> title('Original Image', 'FontSize', fontSize);
> set(gcf, 'Position', get(0,'Screensize')); % Maximize figure.
>
> % Blur the image with a 5 by 5 averaging (box filter) window.
> blurredImage = conv2(grayImage, ones(5,5)/25);
> subplot(2, 2, 2);
> imshow(blurredImage, []);
> title('Blurred Image', 'FontSize', fontSize);
>
> % Perform a variance filter.
> % Output image is the variance of the input image in a 3 by 3 sliding
> window.
> VarianceFilterFunction = @(x) var(x(:));
> varianceImage = nlfilter(grayImage, [3 3], VarianceFilterFunction);
> % An alternate way of doing the variance filter is on the next line:
> % varianceImage = reshape(std(im2col(originalImage,[3 3],'sliding')),
> size(originalImage)-2);
> subplot(2, 2, 3);
> imshow(varianceImage, [])
> title('Variance Image', 'FontSize', fontSize);
>
> % Compute the square root of the variance image to get the standard
> deviation.
> standardDeviationImage = sqrt(varianceImage);
> subplot(2, 2, 4);
> imshow(standardDeviationImage, [])
> title('Standard Deviation Image', 'FontSize', fontSize);
> elapsedTime = toc(startTime);
>
> message = sprintf('Done!\n\nElapsed time = %.2f seconds.',
> elapsedTime);
> msgbox(message);
> return; % End of local_variance() function.

Great code!

However I don't have "imshow' or "nlfilter", i.e. I don't have image
processing toolbox... and lets not get into that procurement etc.

I am thinking of maybe there is a generic way in Matlab of doing
"recursive" computation,

be it recursive mean, or recursive variance/std, or other moments,
etc.

Any thoughts? Thanks a lot!
From: Steve Amphlett on 5 Jul 2010 12:48
Luna Moon <lunamoonmoon(a)gmail.com> wrote in message <6fc04321-2ece-4f41-ad2d-4a0a2b436bae(a)x27g2000yqb.googlegroups.com>...
>
> Great code!
>
> However I don't have "imshow' or "nlfilter", i.e. I don't have image
> processing toolbox... and lets not get into that procurement etc.
>
> I am thinking of maybe there is a generic way in Matlab of doing
> "recursive" computation,
>
> be it recursive mean, or recursive variance/std, or other moments,
> etc.
>
> Any thoughts? Thanks a lot!

Yes. Why do you use the word "recursive"? Your inputs do not depend on your outputs.
From: ImageAnalyst on 5 Jul 2010 12:49
Well just use blockproc and the var() functions - that's one way.
 |  Next  |  Last
Pages: 1 2
Prev: How do you do a moving median fast?
Next: How to align dates in Matlab fast?