| 摘 要: 压缩感知理论是一种全新的数据采集技术,其采用非自适应线性投影来保持信号的原始结构,通过数值最 优化问题准确重构原始信号。本文利用压缩感知的优秀特性,采用基于稀疏表示的模式分类方法,通过提取红外人脸图像 的全部信息作为特征并建立特征矩阵,将待识别人脸作为压缩感知测量值,并通过正交匹配追踪算法进行重构,根据重构 的稀疏系数所属类别进行红外人脸识别。实验表明,基于压缩感知的红外人脸识别结果准确率高。实验验证了本算法的 有效性。 |
| 关键词: 压缩感知；稀疏表示；红外人脸识别 |
|
中图分类号: TP751.1
文献标识码: A
|
| 基金项目: 本文系沈阳师范大学科研基金资助项目“无线图像压缩中感兴趣区域保护及抗干扰技术研究”(L201516). |
|
| Infrared Face Recognition Based on Compressed Sensing |
|
DU Mei,CAO Weiran
|
( Software Institute, Shenyang Normal University, Shenyang 110034, China)
|
| Abstract: As a new data acquisition technology,compressed sensing theory uses non-adaptive linear projection to maintain the original structure of the signal,and accurately reconstructs the original signal through numerical optimization. In this paper,a pattern classification method based on sparse representation is used.By extracting all the information of the infrared face image as the features and establishing feature matrix,the face to be recognized is taken as compressed sensing measurement value,which is reconstructed through the orthogonal matching pursuit algorithm.Finally,the face is recognized according to the category of the reconstructed sparse coefficient.Experiment results prove the high accuracy and effectiveness of the infrared face recognition based on compressed sensing. |
| Keywords: compressed sensing;sparse representation;infrared face recognition |
