Abstract

Human skin detection and face detection are important and challenging problems in computer vision. The use of color information has increased in recent years due to the lower processing time of face detection compared to black and white images. A number of techniques for skin detection are discussed. Experiments have been performed utilizing deep learning with a variety of color spaces, showing that deep learning produces better results compared to methods such as rule-based, Gaussian model, and feed forward neural network on skin detection. A challenging problem in skin detection is that there are numerous objects with colors similar to that of the human skin. A texture segmentation method has been designed to distinguish between the human skin and objects with similar colors to that of human skin. Once the skin is detected, image is divided into several skin components and the process of detecting the face is limited to these components—increasing the speed of the face detection. In addition, a method for eye and lip detection is proposed using information from different color spaces.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call