欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
基于高分辨距离像的雷达目标识别的关键是如何提取距离像的有效鉴别特征。通过分析核Fisher准则函数下核类内散布矩阵的零空间信息与非零空间信息对识别的贡献,基于同构映射思想,求解出一组最优鉴别矢量集,用于提取距离像的鉴别特征。同时,给出一种快速算法,用较少的训练样本表达投影方向,加快了鉴别特征的提取速度。基于外场实测数据的实验证明了方法的有效性。
英文摘要：
In the radar target recognition based on high resolution range profile （HRRP）,the challenging task is the effective feature extraction of HRRP. The useful discriminant information in the null space of the kernel within—class scatter matrix and in the space of the kernel within—class scatter matrix was analyzed based on the kernel Fisher criterion,and an optimal discriminant vectors was acquired based on the idea of isomorphic mapping for the discriminant feature extraction of HRRP. Further more,in order to improve computational efficiency,a simplifying algorithms was proposed based on the projection vector which may be expressed by a part of training samples. Experimental results using outfield real data demonstrate the validity of the proposed method.
同期刊论文项目
同项目期刊论文
Pattern Expression Nonnegative Matrix Factorization: algorithm and Applications to Blind Source Sepa
期刊信息
位置：