Can I use Image entropy in noise removal algorithms, in order to chech their effectiveness? -

i working in digital image restoration field, have studied number of image noise removal research papers, , of these papers using psnr check effectiveness of algorithms, 1 thing noticed ssim page, psnr, depends on mse and, one weakness of mse that, measure depends on scaling of variables despite fact image invariant scaling..

so question this.

can use image entropy check effectiveness of noise removal method method.

of course can that, see [ http://scholar.google.co.uk/scholar?q=image+denoising+entropy ]

the list shows entropy measure works better in domains others. example: if know optimal basis efficiently represents noise-free image (e.g. fourier basis or wavelet basis) cannot efficiently model noise, transformed noise-free image sparse in transform domain , noise-free image not. , sparse signals have low entropy, while dense signals have high entropy.

if know these things true, can evaluate denoising method using transform-domain entropy measure.

you need work calibrate new error message, , of course cannot use entropy-based methods denoising. double dipping.