未找到相关数据
Face feature and parameter recognition algorithm based on Gabor transform
被引：0
作者：
Liu, Lina
[1
]
Liu, Jiaomin
[2
]
Yang, Dongfang
[3
]
机构：
[1] School of information science and engineering, Yanshan University, Ocean college of Hebei Agricultural University, Hebei Qinhuangdao,066000, China
[2] School of information science and engineering, Yanshan University, Hebei Qinhuangdao,066000, China
[3] Qinhuangdao Institute of Technology, Hebei Qinhuangdao,066000, China
基金：
中国国家自然科学基金;
关键词：
Gabor filters - Template matching - Wavelet transforms - Classification (of information);
D O I：
暂无
中图分类号：
学科分类号：
摘要：
The facial expression features are extracted by two-dimensional Gabor wavelet transform, and the elastic template matching algorithm is used to identify facial expressions in the image. In this paper, a face recognition algorithm based on two-dimensional Gabor wavelet is proposed. The algorithm first performs multi-resolution Gaubr wavelet transform on the face image, and then places a set of grid nodes on the image. Aiming at the illumination change and attitude problem in face recognition, it is proposed to first obtain the face image that has been normalized. Firstly, it is preprocessed, then Gabor wavelet transform is performed on the expression sub-region to extract the expression feature vector, and then the expression elastogram is constructed. Finally, the features of the basic expressions extracted by different testers under different illumination conditions are analyzed and compared. ICA method is used to extract independent component, and nearest neighbor classifier is used to classify the feature. A large number of experiments are carried out on ORL face database. The algorithm can get a high recognition rate with a small number of training samples, and has a good adaptability to illumination, facial expressions and other changes. © 2018 Indian Pulp and Paper Technical Association. All Rights Reserved.
引用
收藏
页码：403 / 413
相关论文