摘要： 针对人脸识别方法在复杂环境下识别性能下降、单一特征表述能力有限的问题,基于梯度方向直方图(HOG)和局部二值模式(LBP)特征,提出一种人脸识别方法。通过提取人脸的HOG特征和LBP特征,利用主成分分析和线性判别分析方法进行线性降维,给出基于加权的特征融合策略。在环境复杂的人脸数据库上进行实验,结果表明,相比于单一的局部特征,该特征融合方法能提高人脸识别的精度与速度。

关键词: 人脸识别, 梯度方向直方图特征, 局部二值模式特征, 特征融合, 识别率

Abstract: Due to complex environment of face recognition and limited express ability of single feature,a face recognition method based on Histogram of Oriented Gradient(HOG) and Local Binary Pattern(LBP) feature fusion is proposed.HOG and LBP features are extracted,and Principal Component Analysis(PCA)and Linear Discriminant Analysis(LDA) method is used to dimensionality reduction.Detection results are captured by feature fusion strategies.Experimental results on the face database show that the feature fusion method can not only get better recognition performance but also improve recognition speed compared with single feature.

Key words: face recognition, Histogram of Oriented Gradient(HOG) feature, Local Binary Pattern(LBP) feature, feature fusion, recognition rate

中图分类号: 

• TP391.41