计算机工程

• 人工智能及识别技术 • 上一篇    下一篇

于均方差度量分块的自动加权稀疏表示算法

魏明俊 a,许道云 a,徐梦珂 b   

  1. (贵州大学 a.计算机科学与技术学院;b.理学院,贵阳 550025)
  • 收稿日期:2016-06-12 出版日期:2017-05-15 发布日期:2017-05-15
  • 作者简介:魏明俊(1991—),男,硕士研究生,主研方向为人脸识别;许道云(通信作者),教授、博士生导师;徐梦珂,硕士研究生。
  • 基金项目:
    国家自然科学基金(61262006,615400500);贵州省重大应用基础研究项目(黔科合JZ字[2014]2001号);贵州省科技厅联合基金(黔科合LH字[2014]7636号)。

Automatic Weighting Sparse Representation Algorithm Based on Blocks by Mean Square Deviation

WEI Mingjun  a,XU Daoyun  a,XU Mengke  b   

  1. (a.College of Computer Science and Technology; b.College of Science,Guizhou University,Guiyang 550025,China)
  • Received:2016-06-12 Online:2017-05-15 Published:2017-05-15

摘要: 人脸重要特征部位所在分块应具有更大的分类表决权,而传统图像分块算法往往忽略该问题。为此,提出一种自动加权稀疏表示算法。通过引入一个带重叠的滑动窗口计算分块像素点的均方差,并给出自动加权策略,对每个分块在最终分类中的权重进行度量。在公共数据集上的实验结果表明,与常用的分类算法及分块算法相比,该算法无论是在对最小残差法还是投票法进行加权时,均能提高识别准确率。

关键词: 局部特征凸显, 均方差度量, 分块加权, 稀疏表示, 人脸识别

Abstract: Image segmentation algorithm ignores the image patch which contains the important local facial feature should have bigger weights of voting in the final classification.To solve this problem,this paper proposes an automatic weighting sparse representation algorithm.Through using sliding windows with overlapping,the mean square error value is calculated,and an automatic weighting strategy is given to measure the weight of each patch in the final classification.Experimental results on the public data set show that,compared with the commonly used classification algorithm and block algorithm,the proposed algorithm can get higher recognition rate in face recognition wherever it gives weights to the minimum error or the biggest votes.

Key words: local feature highlight, mean square deviation, block weighting, sparse representation, face recognition

中图分类号: