基于扰动方法的广义最佳鉴别矢量集求解的一种迭代算法

本文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 ,改进了郭提出的求解广义最佳鉴别矢量的一种迭代算法 ,提出了求解广义最佳鉴别矢量的一种新的迭代算法 .本文算法的一个突出优点是随着类别数目的增加 ,计算时间反而缩短 ;而老算法随着类别数目的增加计算时间随着增加 ;不仅如此 ,新算法的识别率不劣于老算法 .在 ORL人脸数据库的数值实验 ,验证了上述论断的正确性

A study has been made on the algorithm of solving generalized optimal set of discriminant vectors in this paper.A new iterative algorithm of solving generalized optimal set of discriminant vectors is proposed on the basis of the improvement of the algorithm by Guo with perturbation theory of matrix.The most obvious advantage of the proposed algorithm is that the computation time decreases when the number of classes increases.However,the characteristics of the conventional algorithm is that the computation time increases when the number of classes increases.Further more,the recognition rate of the proposed algorithm is not less than that of Guo's algorithm under most circumstances.These statements are supported by the numerical simulation experiments on facial database of ORL.

国家自然科学基金课题 (编号:60 0 72 0 3 4)资助;;中国科学院沈阳自动化研究所机器人学研究室基金资助 (编号 :RL2 0 0 10 8);;江苏省自然科学基金课题 (BK2 0 0 2 0 0 1)资助 ;;江苏省高校自然科学研究计划项目(编号 :0 1KJB5 2 0 0 2 )的资助