Signal Dependent Noise Level Estimation

Generalized signal dependent noise model:
y = f + f^gamma*u + w
This code will estimate the three parameters gamma, sigma_u, sigma_w of the Generalized signal dependent noise model.
% Inputs:
% img - color or grayscale noisy image.
% patchsize(optional) - patch size, default 7.
% conf(optional) - confidence interval to determin the threshold for the weak texture.
% In this algorithm, this value is usually set the value very close to one. (default: 1-1E-6).
% itr (optional) - number of iteration. (default: 3).

% Outputs;
% sigw - sigma_w.
% sigu - sigma_u.
% gamma - gamma.
% BW - indices of selected patches.

Reference:
1. Estimation Of Signal Dependent Noise Parameters From a Single Image
Xinhao Liu, Masayuki Tanaka and Masatoshi Okutomi
Proceedings of IEEE International Conference on Image Processing (ICIP2013), September, 2013
2. Practical Signal Dependent Noise Parameter Estimation From A Single Noisy Image
Xinhao Liu, Masayuki Tanaka and Masatoshi Okutomi
IEEE Transactions on Image Processing, Vo.23, No.10, pp.4361-4371, October, 2014