Human facial feature detection based on skin color and edge labeling

Abstract

In the recent years facial feature detection and recognition is an interesting and challenging work in the field of image processing in which facial region detection is the most basic step. This paper proposes a framework for detecting the facial region and facial features from a frontal facial image. Initially, an RGB facial image is used to detect the facial region. In the next step, the facial features, i.e. eyebrows, eyes, nose, and lips regions are detected from the extracted facial region. For that, the binary information is used to detect the facial region from the input facial image. After that, the RGB image is converted into YCbCr based skin color image from where the edges of the facial features are extracted. In the next step, labeling operation is performed on the edge image to identify the location of the facial features region. Various frontal facial images are utilized to evaluate the proposed framework and presented outcomes demonstrate the adequacy.