广义统计不相关最优鉴别矢量集的一个理论结果

本文对统计不相关最优鉴别矢量集的理论问题进行研究 ,提出了广义统计不相关最优鉴别准则 ,并给出了广义统计不相关最佳鉴别矢量集的一个理论结果 ,研究表明 ,广义统计不相关最佳鉴别矢量集的计算公式与基于Fisher最优鉴别准则的统计不相关最佳鉴别矢量集的计算公式完全一样 ,但是以前这一点没有被认识到 .本文的研究丰富了统计不相关最优鉴别分析的特征抽取理论 .

This paper presents a research on the theoretical issue of optimal set of statistically uncorrelated discriminant vectors.First,the criterion of the generalized optimal set of statistically uncorrelated discriminant vectors is proposed,then a theoretical result on the generalized optimal statistically uncorrelated discriminant vectors is obtained.The research result indicates that the formula of the generalized optimal set of statistically uncorrelated discriminant vectors is the same as that of the conventional optimal set of statistically uncorrelated discriminant vectors,however,which was not realized.The research enriches the theory of feature extraction based on the statistically uncorrelated discriminant analysis.

国家自然科学基金课题 (No.60 0 72 0 34) ;;中国科学院沈阳自动化研究所机器人学研究室基金 (No .RL2 0 0 1 0 8) ;;江苏省高校自然科学研究计划项目 (No .0 1KJB52 0 0 2 );;江苏省自然科学基金 (No.BK2 0 0 2 0 0 1 ,BK2 0 0 4 0 58) ;;图象处理与图象通信实验