A HUMAN FACE PROFILE IDENTIFICATION SYSTEM USING 1-D REAL FOURIER DESCRIPTORS

Over the last twenty years, researchers in the field of human face rec ognition have tried to answer the question, how can a face be formally described? Each new face recognition system provides a different answ er. In this paper, the face profile is represented using a real Fourie r descriptor. Outlines of face profiles are extracted from back-lit im ages captured using a CCIR standard TV camera and frame-grabber. Point s on the nose tip and chin are extracted and used to standardise the o rientation and scale of the profile. The face curve is sampled uniform ly in the vertical direction and transformed from a 2-D to a 1-D repre sentation. After Fourier transformation, a small number of the resulti ng coefficients are used as a feature vector. This descriptor is insen sitive to translation, scale and orientation of the original face prof ile. Applications to face recognition and reconstruction are illustrat ed and discussed in the paper.