Center-symmetric local binary patterns (CSLBP)

This computes the Histogram of CSLBP operator, which is an extension of LBP. Using CSLBP is really efficient for illumination and blur type of image transformation. It returns the unnormalized CSLBP histogram of length 16. One can easliy normlize as per his application.
Mostly, it is used as keypoint descriptor. Detect the keypoints, estimate the local patch around the keypoint and then compute the CSLBP descriptor. However, for the sake of simiplicity, below example compute the image level descriptor instead local keypoint descriptor.
Example:

%% Load image
img = imread('myImage.jpg');
%% conver to gray scale if RGB
img = rgb2gray(img);
h = CSLBP(img); % it returns the CSLBP histogram of given image/patch.

One can easily normalize the vector between 0-1 as suggested in original paper
h = h./sum(h);