融合Gabor、LBP与LPQ特征的面部表情识别

Facial Expression Recognition Based on Gabor, LBP and LPQ Features

针对传统的Gabor无法兼顾识别率与实时性的缺点,提出了一种融合Gabor、LBP、LPQ三种特征的表情识别算法。首先采用Gabor变换提取人脸图像的边缘信息,根据获得的变换表征结果,提取其LBP特征及LPQ特征;通过PCA算法对提取的特征进行降维,并对降维后的LBP特征及LPQ特征进行直方图操作;最后,设计基于ELM神经网络面部表情分类器。应用JAFFE人脸表情数据库的测试结果表明,该方法比传统方法具有更高的识别准确度和更快的识别速度。

In order to overcome the shortcomings of the traditional Gabor, which can not take into account both the recognition rate and real-time, a facial expression recognition algorithm fusing Gabor, LBP, LPQ is proposed. Firstly, Gabor transform was used to extract the edge information of the face image, and the LBP and LPQ features were extracted according to the results of the transform characterization. The PC A algorithm was used to reduce the dimension of the extracted characteristics, and extract the histogram of the feature. Finally,the facial expression classifier based on ELM neural network was designed. The test results of JAFFE facial expression database show that the proposed method has higher recognition accuracy and faster recognition speed than traditional methods.