基于核方法的二维线性判决分析的人脸识别算法

Kernel method in face recognition based on 2DLDA

针对基于二维线性判决分析的人脸识别算法中缺少非线性判决信息的问题,提出了一种改进的基于核方法的二维线性判决分析的人脸识别算法。实验结果表明,改进后的算法相对原算法具有更好的识别效果。在此基础上研究了在使用多项式核函数时本文算法的性能,得出了在选用低次数多项式核函数时识别率较高的结论。

The face recognition method based on two dimensional discriminant algorithm lacks the use of nonlinear feature. To overcome this shortcoming, a two dimensional diseriminant analysis algorithm based on kernel method is proposed for face recognition. Test results show that the proposed algorithm performs better than the original one. Further research on polynomial kernel functions shows that the lower degree polynomials receive higher recognition rate.