文档介绍：Pattern Recognition 41 (2008) 3842 --3859
Contents lists available at ScienceDirect
Pattern Recognition
journal homepage: ate/pr
Exploiting Voronoi diagram properties in face segmentation and feature extraction
Abbas Cheddada,∗, Dzulkifli Mohamadb, Azizah Abd Manafc
aSchool puting and Intelligent Systems, Faculty puting and Engineering, University of Ulster, Northern Ireland BT48 7JL, UK
bFaculty puter Science and Information System, University of Technology Malaysia (UTM), Johor, Malaysia
cFaculty puter Science, Malaysian Military Academy (ATMA), Kem Sungai Besi, Kuala Lumpur, Malaysia
ARTICLE INFO ABSTRACT
Article history: Segmentation of human faces from still images is a research field of rapidly increasing interest. Although
Received 28 January 2008 the field encounters several challenges, this paper seeks to present a novel face segmentation and facial
Received in revised form 15 April 2008 feature extraction algorithm for gray intensity images (each containing a single face object). Face location
Accepted 9 June 2008 and extraction must first be performed to obtain the approximate, if not exact, representation of a given
face in an image. The proposed approach is based on the Voronoi diagram (VD), a well-known technique in
Keywords: computational geometry, which generates clusters of intensity values using information from the vertices
Biometric of the external boundary of Delaunay triangulation (DT). In this way, it is possible to produce segmented
Voronoi diagram image regions. A greedy search algorithm looks for a particular face candidate by focusing its action in
Delaunay triangulations elliptical-like regions. VD is presently employed in many fields, but researchers primarily focus on its
Distance transformation use in skeletonization and for generating Euclidean distances; this work exploits the triangulations (.,
Features extraction Delaunay) generated by the VD for use in this field. A distance transformation is applied to