FACE RECOGNIZING AND FACE TRACKING METHOD USING RADIAL BASIS FUNCTION NEURAL NETWORKS (RBFNN) PATTERN CLASSIFIER AND OBJECT TRACKING ALGORITHM AND SYSTEM FOR EXECUTING SAME

摘要

The present invention relates to a face recognizing and face tracking method using a radial basis function neural networks (RBFNN) pattern classifier and an object tracking algorithm and a system for executing the same. More specifically, the system includes a learning step and a test step performed after the learning step. The learning step includes the following steps: (1) detecting face images of a plurality of poses; (2) pre-processing the face images of the poses individually; (3) allowing data of each pose to be learnt; and (4) obtaining an optimum parameter for the data by each pose. The test step includes the following steps: (a) detecting a test face image; (b) determining a pose most similar to the test face image; (c) recognizing the test face image; (d) and tracking the test face image.;COPYRIGHT KIPO 2016