中文摘要：

通过Garbor小波提取人脸表情特征,为降低Garbor变换后向量维数和提取有效的鉴别特征,将手动选取特征点和监督局部线性嵌入（SLLE）结合起来,利用人脸表情图像数据本身的非线性流形结构信息和样本标签信息来调整点到点之间的距离,并形成距离矩阵,而后基于被调整的距离矩阵进行线性近邻重建来实现维数约简,提取低维鉴别特征用于人脸表情识别。结果表明该方法能更为有效地提取反映表情状态的特征,识别率优于传统的PCA算法,取得了较好的识别效果。最后实验分析了SLLE算法近邻数K和嵌入维数对识别率的影响,得到了SLLE算法的最优近邻数K和低维嵌入维数。

英文摘要：

By Garbor wavelet,extracting human facial expression characteristics,in order to incorporate semi-supervised learning and manually selected feature points to reduce the vector dimension and extract effective identification features,a new semi-supervised manifold learning for facial expression recognition is proposed.This method relied on the distance matrix formed by labeled samples,and then the local linear embedding（LLE） method is used to extract discriminative manifold features according to the modified distance matrix.The proposed method produces better classification performance which captures the intrinsic manifold structure collectively revealed by labeled samples.Experimental results on public facial expression databases show that the proposed method can improve facial expression classification performance effectively.Recognition rate is superior to the traditional PCA algorithm.Finally experiments analyze the relationship between neighbor K and the embedding dimension of algorithms SLLE to the recognition rate.