Improved level set method for lip contour detection

The lip contour detection is essential in many applications such as personal identification, facial expressions classification, man-machine interaction, and so on. In this paper, we propose a new level set approach to detect the lip contour. The approach optimizes the gradient information and enhances the accuracy of the lip contour detection by combining YCbCr color space and Fisher transformation. The proposed method consists of the following steps: illumination compensation of the input images, face detection, lip region obtaining, gradient information strengthen by Fisher transformation, and lip contour detection using level set methods. Experiments results show that the proposed method can improve the accuracy of detecting the lip contours.