基于纹理特征融合的人脸表情识别  

Facial Expression Recognition Based on Texture Feature Fusion

作　　者：张雪梅 公维宾 邬建志 王超[1] ZHANG Xue-mei;GONG Wei-bin;WU Jian-zhi;WANG Chao(School of Information Engineering,Chang’an University,Xi’an 710064,China)

机构地区：[1]长安大学信息工程学院,陕西西安710064

出　　处：《计算机技术与发展》2020年第3期57-61,共5页Computer Technology and Development

基　　金：国家自然科学基金面上项目(61572083)。

摘　　要：局部二值模式(LBP)和韦伯局部描述算子(WLD)是两种图像的纹理描述算子,在图像的特征提取方面有较强的能力。为了更加准确地对人脸表情进行识别与分类,针对LBP在特征提取的过程中只考虑了中心像素点与周围的其他像素点的灰度值之差,WLD仅考虑中心像素点与周围像素点灰度值之间的激励强度与梯度方向关系的问题,提出一种新的特征提取算法—局部二值韦伯模式(LBWP)。首先对图像进行预处理,检验人脸和裁剪有效的表情区域,接着对图像进行LBWP特征提取,在特征提取之后采用SVM的分类器对表情进行识别和分类。该算法在CK+数据集和JAFFE数据集上进行实验仿真,识别率分别达到了97.14%和95.77%。实验结果验证了LBWP算法在表情识别方面的有效性,且丰富了人脸图像特征提取方法。Local binary pattern(LBP)and Weber local descriptor(WLD)are two kinds of texture descriptors,which have strong ability in feature extraction.In order to recognize and classify facial expressions more accurately,for these problems that LBP only considers the difference of gray value between the center pixel and other surrounding pixels in the process of feature extraction,and WLD only considers the relation between the intensity of excitation and the direction of gradient between the center pixel and the gray value of the surrounding pixel,a new feature extraction algorithm is proposed,which is local binary Weber model(LBWP).Firstly,the image is preprocessed to verify the face and clip effective expression area,and then the image is extracted with LBWP features.After feature extraction,the facial expression is recognized and classified by SVM classifier.The algorithm is simulated on CK+dataset and JAFFE dataset,and the recognition rate reaches 97.14%and 95.77%respectively.Experimental results verify the effectiveness of LBWP in facial expression recognition,and enrich the face image feature extraction methods.

关 键 词：人脸表情识别 局部二值模式 韦伯局部描述符 特征融合 支持向量机 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]