I want to know how to detect shadow with this article

I can't understand many things in this article.

"So next, the non-shadow point’s mean and standard deviations for the sliding window are computed. Now, the pixels that have intensity less than the one standard deviation of the windows are considered shadow pixels. Figure 3 portrays a successful shadow detection process via the proposed framework."

What should I do for this method?

[Jan]

Create a method to apply a sliding window. Calculate the standard deviation of the intensity. If the central pixel of this window has less intensity, it is considered as shadow.

Try it and post the code, even if it is not working yet. Then it is easier to suggest further improvements.

Note that it is unlikely, that the member of the forum will study your paper and implement the algorithm for you.

[sol park]

I calculated the standard deviation of sliding window with "std(sliding_window);". And I compared the central pixel of this window with the standard deviation. But all the pixel values were bigger than the standard deviation.

[sol park]

Please let me know when I use the standard deviation in this article.

Answer 1

[Michaelscheinfeild]

ycbcrmap = rgb2ycbcr(rgbmap);
meany=mean2(ych);

to be cont

Answer 2

[Michaelscheinfeild]

ycbcrmap = rgb2ycbcr(rgbmap); meany=mean2(ych);

% this generates the binary image xbin=(r+g+b)/3 >meany