摘要
在单样本人脸识别中,由于训练样本的数量受限,充分利用样本的信息就显得十分重要。针对这种情况,通过增加样本虚拟信息,提出了基于虚拟信息的单样本分块人脸识别方法,充分利用了样本的整体信息和局部信息。实验表明,在对人脸图像进行识别时取得了较好的效果,在一定程度上提高了对姿态变化及不精确人脸规范化的适应能力。
As the number of training samples is limited in the face recognition with one training sample per person, it is quite important for using the information of the samples. Aimed at this goal, a method for sub-block face recognition based on virtual information with one training image per person is proposed. The method makes full use of the best of the globe and local information of the samples. The experiment shows the proposed method obtains a better result for recognizing face images. To a certain extent, the adaptability with imprecise face standardization is improved.
出处
《数据采集与处理》
CSCD
北大核心
2009年第4期443-448,共6页
Journal of Data Acquisition and Processing
关键词
虚拟信息
特征提取
分块
图像旋转
virtual information
feature extraction
sub-block
image circumgyration
