基于非下采样Contourlet变换的人脸表情识别算法研究

Facial Expression Recognition based on the Next Sampling Contourlet Transform Algorithm Research

本文研究了非下采样Contourlet变换在人脸表情识别中的应用,并设计了相应的算法流程。首先将人脸表情图像分割为最能表征表情信息的眼睛和嘴巴两个部分,然后利用非下采样Contourlet变换对分割的局部图像进行特征提取,最后使用极限学习机进行分类,并与BP神经网络进行对照实验。研究结果显示,表情分类平均准确率可达86.57%,比BP神经网络的分类方法平均准确率高出7.43%。而在执行速度方法,极限学习机却是BP神经网络的11.09倍,表明了本实验方案的高效性和可行性。

This paper studies the next sampling Contourlet transform in the application of facial expression recognition,and the corresponding algorithm design process. The image segmentation of facial expression is divided into two parts covering the eyes and the mouth,which to the most extent can represent the expression information,and then uses the sampling Contourlet transform under the division of local image feature extraction,furtherly using extreme learning machine for classification,and makes the comparison with the BP neural network control experiment. The results showed that expression classification accuracy can reach 86. 57% on average,than the BP neural network classification method of average accuracy higher than 7. 43%. In the speed of execution method,the fact that extreme learning machine is 11. 09 times that of the BP neural network shows the efficiency and feasibility of the experiment scheme.