you can use the inbuilt commands in MATLAB (or) try the formula "plogp". are u using images as input?

Yes images as input .Do you mean by inbuilt commands functions like max()?

I want to know recent methods both research wise and implementation wise too.

Not max. I think you have a command ent

Or you can try this

for i=1 to n

for j=1:n

f= p(i,j) log p(i,j)

end

end

You can also use a compression method called entropic distance, which measures the difference in entropy between two sequences or two images or any two data sets. The method involves compressing a sample of one dataset and a sample of another dataset, using the same compression algorithm, and normalising the results by compressed cross-samples. Refer to my papers: "Entropic Distance for Automatic Modulation Recognition of HF Signals", "Aspects of HF Communications: HF Noise and Signal Features", "Modulation Recognition for HF Signals", and "Modulation Recognition for Real HF Signals". The method is clearly explained in these papers.

I hope this helps.