CAJ | 학술논문

将神经网络集成应用于多表情人脸识别,通过二维主成分分析获得人脸表情特征,并为每一表情的特征空间各训练一个神经网络,利用另一神经网络时其进行集成.实验结果表明,多神经网络集成方法的识别精度高于单一神经网络所获得的结果.
장신경망락집성응용우다표정인검식별,통과이유주성분분석획득인검표정특정,병위매일표정적특정공간각훈련일개신경망락,이용령일신경망락시기진행집성.실험결과표명,다신경망락집성방법적식별정도고우단일신경망락소획득적결과.
Neural network ensemble is applied to expression invariant face recognition.The facial features used age extracted through two-dimension principal component analysis.Several neural networks are trained for an eigenspace of difference expressions respectively,and their results ale combined with another neural network to recognize the test sets.Experimental results show that the recognition accuracy of the proposed approach is better than individual neural network.