Human Face detection using Skin Color Segmentation and Morphological Operations

Author:   Hanaa Mohsin & Salma Hameedi Abdullah

Download PDF

Abstract The face detection and segmentation images play a vital part in the procedures of image processing and computer vision. It is the procedure of separating digital image to the separate parts have a homogeneous features. Aim of segmentation is the extraction of many properties of the image which are utilized to analyze and detect human face. Segmentation is used in different areas such as medical images, shape identification, content-based image retrieving, robotics vision, and so on. Many approaches were enhanced for image segmentation like pixel-based segmentation, edge based segmentation and region based segmentation. In this paper, we used YCbCr color models for skin color segmentation. These color models with thresholds, help to remove non skin like pixels from an image after that the Morphological operation is used to smooth the object boundary without changing the respective area using erosion and dilation. Then detect the region of face that gives maximum correlation with template. The method is applied for 30 different image subject from Head Pose Image Database. Experimental results show that the proposed algorithm is good enough to localize a human face in an image with an accuracy of 95%. Proposed method is able to achieve good detection success rates for near-frontal faces in real time video.

Keywords: Face detection, image Segmentation, Morphology operation, and Hole Filling