Accurate classification of face images is essential for applications that involve automatic face authentication (e.g. security systems, surveillance applications, etc.). The fundamental operation in such applications is face recognition, which involves the computation of similarity between face images.ASU researchers have developed a novel technique for classification of face images that employs Curvature-based Multi-scale Morphology (CMM). Multi-scale Morphology is an image analysis technique that employs elements whose spatial dimensions are scaled successively.This "scale-space" representation of images has proven to be an efficient technique for indexing a large database of images. A majority of the existing techniques for multi-scale morphology employ regular and symmetrical structuring elements like cylinders, hemispheres or circular poweroids. The shape of these structuring elements is controlled only by the scaling parameter. Our technique uses a structuring element whose shape is a function of both the scaling factors and the principal curvatures of the intensity surface of the face image. This technique has proven to be superior to the standard approach in terms of classification performance.
|Original language||English (US)|
|State||Published - Dec 7 2001|