3D face recognition by shadow moiré

This research applies the shadow moiré method to recognize a 3D face. The contour information of a face is stored in a database by a hierarchical structure from outer to inner contours, which contains the 3D features of the face. The contour similarity rate can be detected by the image moment invariant method. This research extends traditional two-dimensional face recognition into three-dimensional analysis to enhance the detection rate of a face.

▶ A shadow moiré method is used to recognize a 3D face. ▶ A low-pass filter is used to reduce the noise of moiré image. ▶ An image enhance method is to get a better contrast image for feature extraction. ▶ Hu's moment unchanged matrix method applies to generate the features of face contour. ▶ 52 people's face images have been tested and the correct recognition rate of testing samples reaches up 97.04%.