对广义最佳鉴别矢量的求解方法进行研究,根据矩阵的扰动理论,改进了作者最近提出的求解广义最佳鉴别矢量的一种解析算法,提出了求解广义最佳鉴别矢量的一种新的解析算法.该算法运行时间短且识别率不劣于老算法,在ORL人脸数据库进行的数值实验,验证了上述论断的正确性.
The algorithm of solving the generalized optimal set of discriminant vectors is studied. A new analytical algorithm of the generalized optimal set of discriminant vectors is proposed on the basis of the improvement of the algorithm proposed by the authors recently with perturbation theory of matrix. The advantages of the new algorithm are that the computational time is short and the accurate recognition rate is high. These statements are supported by the numerical simu lation experiments conducted on ORL face database.