This implementation estimate noise level in an image as specified in paper
entitled as Single-Image Noise Level Estimation for Blind Denoising by
Xinhao Liu, Masayuki Tanaka, and Masatoshi Okutomi.
Function NLEstimate is the main file which perform this task. Detail are as,

INPUTS:
I = Image
ps = Patch Size (Optional); default size in 7
maxiter = Number of iteration (Optional); default value is 5
OUTPUTS:
ENL = Estimated Noise Level Can be a single value if I is
grayscale image or vector of 1 X 3 dimension if I is
RGB image representing in red, green and blue channel

USAGE:
Estimate Noise Level with default value.
ENL = NLEstimate(imread('football.jpg')); % Return Estimated noise level
for all channel in an image, i.e. for red, green, and blue
ENL = NLEstimate(rgb2gray(imread('football.jpg'))); % Return Estimated
noise level.

REFERENCES:
[1] [Xiang Zhu, and Peyman Milanfar] Automatic Parameter Selection for
Denoising Algorithms Using a No-Reference Measure of Image Content
[2] [Xinhao Liu Masayuki Tanaka Masatoshi Okutomi] Noise Level Estimation
Using Weak Texture Patches of Single Noisy Image
[3] [Xinhao Liu, Masayuki Tanaka, and Masatoshi Okutomi] Single-Image
Noise Level Estimation for Blind Denoising

See Also
http://www.mathworks.com/matlabcentral/fileexchange/36921-noise-level-estimation-from-a-single-image