Zeynab Rostami - Department of Computer Engineering, Faculty of EngineeringArak Branch, Islamic Azad University, Arak, Iran
Baharak Heidari - Department of Computer Engineering, Faculty of Engineering, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran

Nowadays, the classification of human skin as a preprocessor has been proposed in the majority of processing systems on different parts of the human skin, such as face detection, face tracking and content-based retrieval of images, pornography videos and human-computer interface systems (HCI). One main challenge to this area is the diversity of color skin in different races, the existence of skin-like areas in imaging scene, illumination intensity changes and noise in image. In this design, the neural network algorithm is used to reduce detection of skin-like and non-skin areas in images. This project is able to separate skin and non-skin areas by a good approximation. The results show good performance of the proposed model with the average accuracy rate 89.05.

skin, non-skin, images, separate, detection