基于局部相位量化与仿生模式的伪装人脸识别算法

伪装条件下的鲁棒人脸识别，目前在人脸识别领域被日益重视，并认为是难点问题之一. 本文采用非伪装建模方法，提出了一种基于局部相位量化特征提取与仿生模式识别理论的伪装人脸识别算法. 该算法采用了局部相位量化方法进行对伪装模式下具有较好鲁棒性的相位统计特征提取，进而采用仿生神经元构建高维几何覆盖形体，有效利用了不同类别人脸特征的连续性，从而避免了伪装模式的干扰. 在AR数据库及采用警用面部复合软件设计建立的伪装数据库上的仿真实验均表明，与现有主流算法相比较而言，本文所提识别算法在伪装条件下取得了较高的识别性能. 

Disguised Face Recognition Algorithm Based On Local Phase Quantization and Biomimetic Pattern

Disguised face recognition (FR) is considered as one of the difficult and important problems in FR field. Rather than disguised modeling, a disguised face recognition algorithm based on local phase quantization (LPQ) feature and biomimetic pattern recognition (BPR) theory is presented in this paper. The LPQ method is applied to extract the phase statistics feature which is robust to the disguised mode and the biomimetic hyper sausage neuron is adopted to construct high dimensional geometry coverage of different classes, which makes full use of continuous characteristics of different class face features while avoids the interruption of the disguised mode. Experiments on the AR database and the disguised face recognition database established by police face combination software show that, compared with the state-of-the-art method, the proposed recognition algorithm can achieve high recognition rate under disguised conditions.