低分辨率人脸序列的超分辨率重构识别  

Face super-resolution reconstruction and recognition from low-resolution image sequences

作　　者：张地[1] 何家忠[1] 彭宏[2] 

机构地区：[1]韶关学院计算机系,广东韶关512005 [2]华南理工大学计算机学院,广东广州510641

出　　处：《韶关学院学报》2007年第6期32-35,共4页Journal of Shaoguan University

基　　金：中国博士后基金资助(20060400218)

摘　　要：针对低分辨率人脸序列的识别这一问题,提出了一种超分辨率重构识别算法.该算法利用低分辨率序列中所提供的关于同一人脸的不同信息,先重构出一幅具有更高分辨率的人脸图像,然后再进行基于Gabor特征的人脸识别.实验结果表明,该算法能够显著提高人脸识别率.The recognition rate of normal face recognition algorithms drops dramatically when the resolution of face is too low.In this paper,a new hybrid face recognition algorithm is proposed to tackle the problem of face recognition from low resolution image sequences.The proposed approach first constructs a highresolution face image by fully utilizing the sub-pixel information provided in the low-resolution sequences,and then performs Gabor feature based recognition.Experimental results are also provided to demonstrate the effectiveness of the proposed algorithm.

关 键 词：超分辨率 图像重构 人脸识别 

分 类 号：TN911.73[电子电信—通信与信息系统]