Improved lip detection algorithm based on region segmentation and edge detection

Lip detection is an important area of research in image processing that has numerous applications. In this paper an improved algorithm for lip detection is proposed. It is based on skin detection using combination of HSV, YCbCr and normalized RGB color models, face detection using largest connected skin region and lip detection after lower part extraction by edge information. Experimental results show that our proposed algorithm, tested on standard benchmark images, exhibits excellent performance and has capacity for improvement.

I. Introduction

Human face perception is currently an active computer vision research topic. Detection of human faces, their locations and face details are prerequisite for face perception. Lip detection is an important branch in this area of research; the significant interest in this topic originates from the numerous applications where lip detection either serves as a preprocessing step or directly provides information for higher level image analysis.