radar HRRP target recognition method based on priori optimal variational autoencoder

Abstract

The invention discloses a radar high-resolution range profile target recognition method based on a priori optimal variational autoencoder, thereby solving a problem of low identification rate in the prior art. The method comprises: step one, acquiring radar high resolution range imaging data and classifying the imaging data into a training sample set and a testing sample set; step two, preprocessing radar high resolution range profile data and constructing a priori optimal variational autoencoder formed by two perceptrons; step three, training the priori optimal variational autoencoder by using the training set data to obtain a trained first multi-layer perceptron and a trained second multi-layer perceptron; step four, extracting a hidden variable feature a corresponding to the training set and training an SVM; and step five, extracting a hidden variable feature b corresponding to the testing set and completing target identification of the testing set by using the trained SVM. Therefore, the identification rate is improved obviously; and the robustness of the noisy sample identification is enhanced. The method can be applied to environmental detection and track tracking.

Description

technical field [0001] The invention belongs to the field of radar technology, in particular to a radar high-resolution range image HRRP target recognition method, which can be used for environment detection and track tracking. Background technique [0002] The range resolution of the radar is proportional to the received pulse width after the matched filter, and the range unit length of the radar transmitted signal satisfies: ΔR is the distance unit length of the radar transmitting signal, c is the speed of light, τ is the matching received pulse width, and B is the bandwidth of the radar transmitting signal. The height of the radar range resolution is relative to the observation target. When the size of the observed target along the radar line of sight is L, if L<<ΔR, the corresponding radar echo signal width is approximately the same as the radar emission pulse width. Usually called "point" target echo, this type of radar is a low-resolution radar; if L>>ΔR,...

AI Technical Summary

Problems solved by technology

[0005] At present, many target recognition methods for high-resolution range image HRRP data have been developed, including using more traditional support vector machines to directly classify targets, using time-domain convolutional neural networks for classification, and using restricted Boltzmann machines. feature extraction methods, but the common deficiency of these methods is: the accuracy of target recognition is not high

Images