【摘要】 针对人脸识别中区域的高阶隐藏非线性结构发现问题,提出了字典学习优化结合2D Krawtchouk矩(KCM)选择的人脸识别方法.首先,利用二维KCM选择提取特征向量;然后,利用字典学习优化得到最优特征矩阵,并将特征进行组合;最后,使用最近邻分类器完成分类.实验结果表明,相比其他几种方法,该方法获得的平均精度高且鲁棒性更好.更多还原

【Abstract】 For the problems of capturing region-based higher-order hidden nonlinear structures on face recognition,a new face recognition method based on dictionary learning optimization and 2D KCM is proposed.Firstly,2D KCM selection is used to extract feature vectors.Then,dictionary learning optimization is used to get the optimal matrix,and features are combined.Finally,nearest neighbor classifier is used to finish the classification.The experimental results show that the proposed method has higher average precision and better robustness than several other methods.更多还原