Face Detection By Finding The Facial Features And The Angle of Inclination of Tilted Face

Face is a popular topic of research in computer vision, image analysis and understanding. This paper proposes a method to detect straight posed face as well as tilted face in colour images. Algorithm first detects skin colour region and then skin region is further processed by finding connected components and holes. Each connected components is tested to extract facial features and Y-value of each holes. If features exist in the connected component then the component is concluded as face. The Y-value of both eyes are compared to find whether the face is straight or tilted. For tilted face, the angle of tilt is determined and the input image is rotated by the angle of tilt to straighten up the face. After straightening the face, the rectangle is drawn to enclose the face. The dimension of drawn rectangle is obtained by using the distance parameter of the line between two eyes.