MODEL COMBINATION ZERNIKE MOMENT AND ARTIFICIAL NEURAL NETWORK FOR DETECTING THE EDGE OF THE FACE WITH EMOTIONS IN THE PICTURE

Abstract

“Detecting the edge of the face in the picture” is an important phase in human face recognition system, the accuracy of detecting the face position will determine the system's performance. To solve this problem, so far there have been many techniques proposed and has gained certain successes, such as AdaBoost, Artificial neural network, SVM. Within the scope of the paper, we propose a model combining zernike moment and artificial neural networks with multi-layer perceptron to solve the problem (their model I call ZM-ANN). Model proposed exploiting the orthogonality and invariance for rotations of geometric Zernike moment.