Radar object distance image recognition method based on kernel discrimination local tangent space arrangement

Abstract

The invention discloses a radar object distance image recognition method based on kernel discrimination local tangent space arrangement, and the method comprises the following steps: obtaining a plurality of training distance images in a known target type and test distance images in a to-be-tested target type; carrying out the preprocessing of the obtained training distance images and the test distance images; obtaining a mapping matrix V from a high-dimensional distance image space to a low-dimensional feature space based on the rules of local tangent space reconstruction error minimization and inter-class scattering maximization; extracting the features of the training distance images and the test distance images: y=VTk, wherein VT is the transposition of the mapping matrix V, and k is the kernel vector of one of the training distance images or the test distance images; comparing the features of the training distance images and the test distance images through employing a nearest neighbor method, enabling each test distance image to be divided into the target class where the nearest training distance image belongs.

Description

technical field [0001] The invention relates to the field of radar target recognition, in particular to a radar target range profile recognition method based on kernel discrimination local tangent space arrangement. Background technique [0002] At present, many methods related to manifold learning theory have been successfully applied in the field of radar target recognition. Local Tangent Space Arrangement (LTSA) is one of the classic manifold learning methods, which directly obtains the low-dimensional projection of high-dimensional data sets. Coordinates, when a new sample arrives, only the new sample is added to the old sample set, and then the new low-dimensional projection coordinates are recalculated, which greatly limits its application in the field of target recognition. [0003] In response to the shortcomings of LTSA, a linear local tangent space arrangement (LLTSA) method was proposed, which still has common defects with LTSA. They only focus on retaining the lo...

AI Technical Summary

Problems solved by technology

[0002] At present, many methods related to manifold learning theory have been successfully applied in the field of radar target recognition. Local Tangent Space Arrangement (LTSA) is one of the classic manifold learning methods, which directly obtains the low-dimensional projection of high-dimensional data sets. Coordinates, when a new sample arrives, only the new sample is added to the old sample set, and then the new low-dimensional projection coordinates are recalculated, which greatly limits its application in the field of target recognition

[0003] In response to the shortcomings of LTSA, a linear local tangent space arrangement (LLTSA) method was proposed, which still has common defects with LTSA. They only focus on retaining the local manifold structure between samples, while ignoring the category information of samples. This will affect the recognition performance to a certain extent

Furthermore, LLTSA is a linear method and has very limited learning capabilities for data with typically nonlinear characteristics such as radar target high-resolution distances

[0004] For nonlinear problems, the kernel method is undoubtedly a good solution. Classic kernel methods such as KPCA and GDA have been successfully applied to radar target high-resolution range image recognition, but both GDA and KPCA use global learning methods. The local structural features between samples are ignored, and their ability to solve the problem of pose sensitivity of high-resolution range images is very limited

Images