Skin-Detection-Naive-Bayes-

Skin detection detection using naive Bayes classification (python)

NOTE:

• The code is for demonstration only, python codes are not optimized in function calls
• Used Naive Bayes to classify skin probable colors and non skin colors
• First all pixels (256x256x256) in RGB, is taken to extract their frequencies as skin color or non skin color
• ^ this is done by comparing mask photos to their original photos
• Then each pixel value [r,g,b] is mapped to its probability of being a skin color
• Finally test any image to extract its skin as white color

RUN:

• Run gather.py (Build a file with probabilitys for each pixel in rgb plane)
• Run test.py

NEED:

• pip install pillow
• python > 3.5

CREDIT:

• https://github.com/MinhasKamal/SkinDetector