REGION-BASED TEMPLATE DEFORMATION AND MASKING FOR EYE-FEATURE EXTRACTION AND DESCRIPTION

We propose an improved method for eye-feature extraction, descriptions , and tracking using deformable templates. Some existing algorithms ar e exploited to locate the initial position of eye features and then de formable templates are used for extracting and describing the eye feat ures. Rather than using original energy minimization for matching the templates, the region-based approach is proposed for template deformat ion. Based on the region properties, the new strategy avoids problems such as template shrinking, adjusting the weights of energy terms, fai lure of orientation adjustment due to some exceptional cases. Our stra tegies are also coupled with Canny edge operator to give a new back-en d processing. By integrating the local edge information from the edge detection and the global collector from our region-based template defo rmation, this processing stage can generate accurate eye-feature descr iptions. Finally, the template deformation process is applied to track ing eye features. (C) 1997 Pattern Recognition Society.