一种改进的LDB人脸识别方法

人脸识别是模式识别领域的一个重要研究课题,它具有广泛的应用背景并日益受到学术界、企业界、政府和军事部分的高度重视。人脸识别研究的目标主要有两个,一是提高识别正确率,二是降低训练与识别时间。论文对传统LDB方法进行改进,基于差可分度确定小波包分解子带,以选定子带内选定系数的一、二阶原点距作为人脸特征,定义了相应的分类识别距离,在此基础上提出了一种新的人脸识别方法,既减少了计算复杂度,降低训练与识别时间,保证实时性,又能够更好地描述对分类有用的人脸特征,提高识别正确率。

Improved LDB method of face recognition based on moment property

Face recognition is an important embranchment of pattern recognition problem.It has been widely applied and highly valued by academic field,enterprise field,government and the military day by day.There are two targets in face recognition: enhance recognition correctness and lower training and recognition time.In this paper,the traditional LDB method is improved to be a new face recognition method.The basis' of subbands in Wavelet packet decomposition are choosen based on difference separability.The first and second origin moments of chosen coefficients in chosen subbands are used as the feature of faces and the classify recognition distance is defined correspondingly.This new method reduces the calculation complexity,lowers the training and recognition time and achieves real time recognition,while the useful face features are better described for classification and recognition correctness are enchanced.