Zernike moments for facial expression recognition

This study presents a facial expression recognition system using an orthogonal invariant moment namely Zernike moment (ZM) as a feature extractor and LDA classifier. Changes in illumination condition, pose, rotation, noise and others are challenging task in pattern recognition system. Simulation results on Cohn-Kanade database show that higher order ZM features are obtained good results in images with noise and rotation whereas feature extraction time rate is slower than other methods.