基于卷积神经网络的人脸表情识别研究

Research on Facial Expression Recognition Based on Convolutional Neural Network

人脸表情识别是图像识别的一个重要领域。由于人脸表情较多,图像背景复杂,不同类型人脸相似,同类型人脸的差异甚小,人脸表情识别仍存在很大挑战。传统人脸识别分类方法主要基于依靠人工提取分类特征,且精度不高。本文构建基于Keras的卷积神经网络模型,并运用FER2013数据集,结果表明该方法提高了人脸表情识别的精度,为该问题的解决提供了新的思路和方法。

Facial expression recognition is an important field of image recognition.Face expression recognition still faces great challenges because of the large number of facial expressions,the complex image background,the similarity of different types of faces and the small difference between the same types of faces.Traditional face recognition classification methods mainly rely on manual extraction of classification features,and the accuracy is not high.This paper constructs a convolutional neural network model based on Keras and applies it to FER2013 data set.The results show that this method improves the accuracy of facial expression recognition and provides a new idea and method for solving this problem.