how to write equation in matlab

please can some help me write the equations below in matlab.... in the following link....

http://img545.imageshack.us/img545/8454/equations.png

please do reply.....

Answer 1

[Walter Roberson]

If I read the equation correctly, then Imu and Isigma do not depend on the x and y parameters and are instead the same for all rows and columns.

Imu = mean2(f)
Isigma = std(f(:), 1);   %the biased standard deviation

[Elysi Cochin]

but sir i need values for each pixel.... if the values are not same.... how would we do it??

[Walter Roberson]

Read the right hand side of the equations. There is a double summation given that varies x in the outer summation, and varies y in the inner summation. Those variation of "x" and "y" by the summation operators override whatever x and y are given as parameters. As the x and y that are the parameters never appear outside of the summations, the result of the calculation depends only on the grayscale values over the sub-array f(1:R, 1:C) where R and C are constants. So Imu is constant for the image rather than being dependent on the x and y parameters.

The same logic applies for Isigma: it too is dependent only on the image contents and independent of the x and y coordinate given as parameters.

The calculations for Imu and Isigma then become as shown above -- mean2() and the biased standard deviation.

But if you insist, you could instead say,

Imu = ones(size(f)) * mean2(f);
Isigma = ones(size(f)) * std(f(:), 1);  %biased standard deviation

Somewhat pointless as you simply replicate the single value to the size of the array, but if it works for you then go ahead and use it.

[Elysi Cochin]

sir actually i wanted to get these values and give as input to neural network..... but this way will it work??? using those equations i'm extracting the features.... and is "f" gray scale image???

[Walter Roberson]

"f" is the grayscale image, just as is indicated in the equation.

[Elysi Cochin]

Isigma = std(f(:), 1); %the biased standard deviation

sir when i do this i get error message

    ??? Error using ==> var at 56
    First argument must be single or double.

    Error in ==> std at 32
    y = sqrt(var(varargin{:}));

    Error in ==> Untitled at 7
    Isigma = std(f(:), 1);

what should i do to rectify error sir....

[Walter Roberson]

Isigma = std(double(f(:)), 1);

[Elysi Cochin]

thank u sir....

[Elysi Cochin]

sir like mean2 function is there any function to calculate "variance"..... is variance and standard deviation the same??? please do reply sir.....

[Walter Roberson]

Isigma = var(double(f(:)), 1);

[Elysi Cochin]

thank u sir...