基于深度卷积神经网络的面部表情识别研究

Facial Expression Recognition Based on Deep Convolutional Neural Network

本研究为了解决传统面部表情识别模型准确率较低的问题,基于深度卷积神经网络(Deep Convolutional Neural Network,DCNN)提出一种新的改进神经网络模型,与传统模型相对比,本模型将其核心的卷积层替换成了深度可分离卷积层,同时搭配卷积残差块的使用,使网络能够有效减少参数的情况下,能够提取多尺度上的特征信息,从而有效地保留了细节特征。最后通过仿真对比,证明本研究提出的DCNN网络具有突出的性能特点,适合用于面部表情识别任务。

The study proposes a new improved neural network model based on Deep Convolutional Neural Network(DCNN)in order to solve the problem of low accuracy of the traditional facial expression recognition model.Compared with the traditional model,this model replaces its core convolutional layer with a depth-separable convolutional layer,and at the same time with the convolutional residual block The use of convolutional residual blocks enables the network to extract feature information on multiple scales with effective parameter reduction,thus effectively preserving detailed features.Finally,through simulation comparison,it is demonstrated that the DCNN network proposed in this study has outstanding performance characteristics and is suitable for facial expression recognition tasks.