Automatic extraction of lips based on multi-scale wavelet edge detection

Various methods for lip segmentation have been proposed, and it still remains a challenging and difficult problem due to the weak colour contrast between the lip and other face regions. A novel automatic lip segmentation algorithm is proposed based on the wavelet multi-scale edge detection across the discrete Hartley transform. Comparative study with some existing lip segmentation algorithms has indicated the superior performance of the developed algorithm. The proposed algorithm produces better segmentation without the need to determine an optimum threshold for each lip image. In contrast to the other methods investigated, the lip segmentation is determined completely automatically.