Skip to content

Skin segmentation is basic and key in the hand gesture recognition. Since skin color is within a threshold range, thresholding a color space, we can swiftly segment the desired region.

Notifications You must be signed in to change notification settings

margaritageleta/human-skin-detector

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Human Skin Segmentation for Hand Gesture classification

With thresholding based on YCbCr, HSV, RBG and CIELab* color spaces intersection and morphological operations in gray scale

Skin segmentation is basic and key in the hand gesture recognition. Since skin color is within a threshold range, thresholding a color space, we can swiftly segment the desired region. However, there are two factors that have a direct influence upon this whole process:

  • Illumination conditions, noise effects and complex backgrounds.
  • Using a particular color space can influence image information representation.

In this paper we propose two alternatives to approach the segmentation problem:

  • histogram-based and
  • color space heuristics.

After successfully segmenting human skin, we train a neural network for hand gesture detection to predict how many fingers are represented by the gesture and the performance is assessed with several metrics with respect to the segmentation hyperparameters.

About

Skin segmentation is basic and key in the hand gesture recognition. Since skin color is within a threshold range, thresholding a color space, we can swiftly segment the desired region.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published