摘要
通过提取人脸图像的Gabor特征,结合Adaboost,进行人脸表情识别(FER)。针对Gabor特征维数高、冗余大的特点,引入Adaboost算法进行特征选择降低特征向量的维数。然后再以支持向量机(SVM)和最近邻分类法相结合组成分类器进行分类。该方法综合运用了Gabor特征对于人脸表情的良好表征能力、Adaboost算法的强大特征选择能力以及SVM在处理少样本、高维数问题中的优势。在JAFFE库上进行测试的结果验证了该法的有效性。从Adaboost所选择的特征集可知,在眼和嘴区域提取的特征,对于FER是最为重要的。
An approach is proposed to recognize the facial expression using Gabor feature and Adaboost. Since the high-dimensional Gabor feature vectors are quite redundant,Adaboost is introduced as a method of features selection. Furthermore,combined with the nearest distance classifier, the support vector machine（SVM） is used for classification. This approach takes the advantages of the favorable ability of Gabor feature in representing expression variability, the effective function of Adaboost in feature selection, and the high performance of SVM in the solution to small sample size, high dimension problems. Experiments with JAFFE show that the approach is quite effective. Meanwhile,the feature set selected by Adaboost also indicates that the features extracted from eye and mouth regions play the most important role in expression recognition.
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2006年第8期993-998,共6页
Journal of Optoelectronics·Laser
基金
国家重点攻关资助项目(2001BA811B07)
