ウェプレット変換とニューラルネットワークを用いた顔表情認識(一般講演,ライフサイエンスとソフトンピューティング)

抄録

This paper relates a method for facial expression recognition based on wavelet transform and neural network. Firstly, the input face image is normalized and is decomposed into wavelet coefficients. Then, the statistical features are extracted from wavelet coefficients. And then, these features are fed to the neuron units in the first layer of a neural network. The neural network is trained for facial expression recognition. We tested this method with the real-world images. Experiment results show that this method is valid for facial expression recognition.