Ground object classification method based on robustness time frequency characteristics

Abstract

The invention discloses a ground object classification method based on robustness time frequency characteristics, which is mainly used for solving the problem that the prior art takes long time to perform classification on the vehicle objects and the human body objects under a low signal to noise ratio condition. The realization process comprises steps of performing normalization on recorded high signal-to-noise ratio signals to obtain training signals, 2 extracting a three-dimension time frequency characteristic and training a classifier from a time frequency spectrum of the training signals, 3 performing normalization on recorded low signal-to-noise ratio signal energy to obtain test signals, 4 extracting 3-dimension time frequency characteristics from the time frequency spectrum of the test signals, 5 transmitting the 3D time frequency characteristics of the test signal into the trained classifier to obtain an classification result. Under the condition that the signal-to-noise ratio is low and has no need of performing beforehand de-noise processing, the ground object classification method based on robustness time frequency characteristics can quickly and effectively realize the classification of the vehicle objects and the human body objects and can be used for radar object identification.

Description

technical field [0001] The invention belongs to the technical field of radar and relates to a target classification method, which can be used to identify ground vehicle targets and human targets. Background technique [0002] In the narrow-band radar system, training data is needed to train the classifier when classifying actual targets. The training data is usually obtained through cooperative experiments or simulation experiments, and the signal-to-noise ratio is high. However, the data used for testing is obtained under non-cooperative environment conditions and cannot Guaranteed to have a high signal-to-noise ratio. Most existing radar target classification methods cannot guarantee effective target classification under the condition of low signal-to-noise ratio. In addition, since human objects and vehicle objects usually serve different tasks, it is of great significance to effectively realize the classification of two types of ground objects under low SNR conditions. ...

AI Technical Summary

Problems solved by technology

Although this method can effectively realize the classification of two types of ground targets under the condition of high signal-to-noise ratio, in order to effectively realize the classification of two types of ground targets under the condition of low signal-to-noise ratio, it is necessary to use the CPPCA denoising method to classify the target The radar echo is denoised first, and the CPPCA denoising method involves matrix inversion operation and eigenvalue decomposition operation, so it increases the calculation amount when classifying two types of ground targets, and affects the classification speed

Images