基于卷积神经网络的人脸表情识别

Facial expression recognition based on convolutional neural network

由于人脸表情之间的差别很细微,为了提高卷积神经网络对人脸表情特征提取能力及其识别精度,把连续卷积引入到卷积神经网络模型中,改进后的模型采用小尺度的卷积核可以使得提取到的人脸表情特征更精密,两个连续的卷积层使模型的非线性表达能力得到增强。提出网络权值优化操作,构建SOM网络进行预学习,将最优学习结果的神经元用于初始化连续卷积神经网络,实验结果表明优化后的卷积神经网络对人脸表情图像识别精度得到了提高。

Because there is a small difference in different facial expressions,In the field of facial expression recognition,in order to improve the recognition accuracy of the convolutional neural network,a convolutional neural network model based on the consecutive convolution was proposed.The small scaled convolutional kernels in the model were adopted to precisely extract the local features,and the nonlinear expression capability of the model was improved with the help of two continuous convolutional layers.The optimization operation of the network weight is proposed,The SOM network was constructed to conduct the preliminary learning for the samples,The learning precision of the network is calculated to initialize the convolutional neural network by means of the neurons with optimal learning result.The experimental result shows that the recognition accuracy of the optimized convolutional neural network is improved.