how to implement these equations in Matlab?

I have the following equations

where and represent the first-order derivative of matrix I . I tried to implement these equations in matlab using the following lines:

   fx = diff(I,1,2);
   fx = padarray(fx, [0 1 0], 'post');
   fy = diff(I,1,1);
   fy = padarray(fy, [1 0 0], 'post');
   Gx = -log10(fx);
   Gy = -log10(fy);

however, the value of Gx and Gy are complex numbers which means that these equations don't implemented correctely. does anyone know how to implement these equations correctely in matlab? any help will be appreciated.

[VBBV]

fx = padarray(fx,[0 1],0,'post');

Similarly for fy. The second argument in pad array is size of matrix. Third arg is value

[ghada sandoub]

thanks for your response. do you mean that the remaining lines of code are correct and this will calculate the logarithmic of derivative correctly?

[VBBV]

May be use natural log instead of decimal log

Gx = -log(fx);

[John D'Errico]

natural log vs log10 is irrelevant. If one gives complex, than so will the other.

The log of a negative number is complex. Taken to ANY base.

[VBBV]

Check if fx and fy Is producing negative values. Also verify the initial conidtions for generating I matrix used as input to diff function