Skin color detection

Computer vision is one of many areas that wants to understand the process of human functionality and copy that process with intention to complete human life with intelligent machines. For better human – computer interaction it is necessary for the machine to see people. This can be achieved by employing face detection algorithms, which algorithm is based on skin color detection. One of the problems this and similar algorithms have to deal with is sensitivity to the illumination conditions under which the input image is captured. Hence illumination sensitivity influences face detection results. One of the aspects from which we can observe illumination influence is the choice of proper color space. Since come color space are designed to eliminate the influence of illumination when describing color of object, an idea of using such a color space for skin-color detection has been taken under consideration.

The skin color for example for fair complexion is determined with the following rules, which describe the skin cluster in the RGB color space:

R > 95 AND G > 40 AND B > 20 AND

max{R, G, B} – min{R, G, B} >15 AND

|R – G| > 15 AND

R > G AND R > B

OR

R > 220 AND G > 210 AND B > 170 AND

|R – G| <= 15 AND

R > B AND G > B

How to teach computers to understand red color?

Why is it so important?

Our world is not so perfect! There are a lot of people who are blind, or whose sight is impaired, and hence can not locate and read attention signs. So there are some technologies what are designed to resolve that problem, such as Global Position System (GPS) and Smith-Kettlewell’s Rehabilitation Engineering Research Center’s “Talking Signs” project. Functionality of those all solutions is to detect, locate and read general information signs, this involves segmenting signs from the background, estimating their pose, and reading them. These are all difficult problems particularly in cluttered street scenes where the illuminant color can vary greatly, where the pose and position are unknown, and where the signs can be partially obsured or in shadow.

Solution!

The color can be represented in different ways, but in this case will be used RGB (Red, Green, Blue) color model. Well the conditions of understanding if some color X is red, are such:

R(X)>128 and G(X)<0.8*R(X) and B(X)<0.8*R(X)

Where R(X) – red value of X color, G(X) – green value and B(X) – blue value