PZM/ANN hybrid network for face recognition

Improved Pseudo-Zernike Moment (PZM) and artificial neural network (ANN) was combined within the hybrid architecture for face recognition. Improved PZM was used to extract face feature, and encoded to form the input vector sending to ANN. Experimental results demonstrate the present approach taking advantage of ANN, basically eliminates the effects of the change of face scale and rotation, and has better robust to variation of illumination, pose and facial expression. On the basis of previous study, the approach makes a further discussion on the application of Pseudo-Zernike Moments in the aspect of face recognition.