In:
Journal of Physics: Conference Series, IOP Publishing, Vol. 1656, No. 1 ( 2020-09-01), p. 012004-
Abstract:
Target classification technology is an important method to improve the operational performance of radar in complex electromagnetic environment, as well as an important means to reduce radar false alarm. In this paper, according to the randomness of target plot parameters, a feature prior model is established to fuse the features distribution of true/false targets based on Bayesian estimation so as to achieve intra-feature information fusion. Under the guidance of the classification hyper-plane, feature dimension reduction is conducted to multi-dimension Gauss distribution after fusion, so as to achieve inter-feature information fusion and obtain 1-D Gauss probability distribution of target features. The target probability is obtained according to the vector distance integral of the target, thus the target classification is completed according to the probability threshold. This paper makes analysis, calculation and simulation to target classification algorithm under the features of 1-D, 2-D, 3-D and multi-dimension so as to validate the effectiveness of the proposed method.
Type of Medium:
Online Resource
ISSN:
1742-6588
,
1742-6596
DOI:
10.1088/1742-6596/1656/1/012004
Language:
Unknown
Publisher:
IOP Publishing
Publication Date:
2020
detail.hit.zdb_id:
2166409-2
Permalink