Automatic Extraction of Lip Based on Wavelet Edge Detection

Abstract

The effective automatic lip extraction from images is fundamental in a variety of applications. Since the outer labial contour of the mouth is very poor color distinction when compared against its skin background, it makes extraction of lip a difficult problem. In order to improve the contrast between lip and the other face regions, discrete Hartley transform (DHT) is proposed. We perform wavelet multi-scale edge detection across the C3 component of the HDT, which takes both the color information and the geometric characteristic into account. Comparative study with some existing lip segmentation algorithms has indicated the superior performance of the developed algorithm. The proposed algorithm produces better segmentation without need to determine an optimum threshold for each face image. In contrast with the other methods investigated, the lip segmentation is determined completely automatically