一种基于Lp范数最小化问题的人脸识别算法  

Face recognition algorithm based on Lp-norm minimization

作　　者：谢晓 王敬 蔡昊 XIE Xiao;WANG Jing;CAI Hao(Xinyang Normal University,Xinyang,Henan 464000,China)

机构地区：[1]信阳师范大学,河南信阳464000

出　　处：《计算机应用文摘》2023年第17期120-122,共3页Chinese Journal of Computer Application

基　　金：信阳师范大学青年科研基金项目(2021-QN-036)。

摘　　要：基于Lp范数最小化问题的稀疏编码是人脸识别、图像重建、压缩感知等领域中的重要方法。文章在迭代重加权最小二乘法的基础上进行改进,提出一种新的求解Lp范数最小化问题的算法。该方法能巧妙地避免除以零的问题,同时将解域拓展到力≥0的范围。最后通过人脸识别实验证明了该方法的有效性。Sparse coding based on Lp-norm minimization is an important method in face recognition,image reconstruction,compressed sensing,and other fields.Based on iteratively reweighted least squares,this paper proposes a new algorithm for solving the Lp-norm minimization problem.The proposed algorithm can elegantly avoid the problem of dividing by zero and expand the solution domain to the range of p≥0.Finally,this paper demonstrates the effectiveness of the proposed method by a face recognition experiment.

关 键 词：稀疏编码 LP范数 人脸识别 迭代重加权最小二乘法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]