Linear discriminant learning true and false target one-dimensional range profile feature extraction method

Abstract

The invention belongs to the technical field of radar target recognition, and particularly relates to a linear discriminant learning true and false target one-dimensional range profile feature extraction method. According to the method, various sample distribution center vectors are obtained through iterative learning; then, a learning center vector is used for replacing a sample mean vector to calculate an intra-class scattering matrix and an inter-class scattering matrix; the linear discriminant learning transformation matrix is obtained through calculation, the degree of intra-class aggregation and inter-class separation can still be well represented under the condition that target sample data distribution is non-Gaussian distribution, the defect that a conventional discriminant subspace is only suitable for sample data Gaussian distribution is overcome, and therefore the target recognition performance is improved.

Description

technical field [0001] The invention belongs to the technical field of radar target recognition, and in particular relates to a linear discrimination learning method for extracting one-dimensional range image feature of true and false targets. Background technique [0002] In radar target recognition, the discriminant vector subspace method can increase the difference between heterogeneous target features while reducing the difference between similar target features, so as to extract effective recognition features. Therefore, the discriminant vector subspace method can obtain good recognition performance. [0003] However, the discriminant vector subspace method is only suitable for the case where the sample data is Gaussian distribution, but in practice the sample data distribution may be non-Gaussian distribution, which leads to the decline of the recognition performance of the discriminant vector subspace method. There is room for further improvement in the recognition p...

AI Technical Summary

Problems solved by technology

[0003] However, the discriminant vector subspace method is only suitable for the case where the sample data is a Gaussian distribution, and the actual sample data distribution may be non-Gaussian distribution, resulting in a decline in the recognition performance of the discriminant vector subspace method

There is room for further improvement in the recognition performance of existing conventional discriminant vector subspace methods

Images