Radar HRRP target recognition method based on spectrogram segmentation preprocessing and convolutional neural network

Abstract

The invention discloses a radar HRRP target recognition method based on spectrogram segmentation preprocessing and a convolutional neural network. The method comprises the steps of firstly, carrying out pretreatment to reduce the sensitivity in an HRRP sample; carrying out spectrogram transformation on the sample; enabling the next CNN layer to extract the information contained in the HRRP envelope while retaining the sequence correlation contained in the HRRP sample; then extracting the high-level features through time sequence correlation modeling by the bidirectional RNN; and finally carrying out target classification through the softmax function.

Description

technical field [0001] The invention belongs to the field of radar target recognition, and in particular relates to a radar HRRP target recognition method based on spectrogram segmentation preprocessing and convolutional neural network. Background technique [0002] The range resolution of high-resolution broadband radar is much smaller than the size of the target, and its echo is also called the one-dimensional high-resolution range profile (HRRP) of the target. HRRP contains extremely valuable structural information for classification and identification, such as the radial size of the target, the distribution of scattering points, etc., and has broad engineering application prospects. Therefore, the radar automatic target recognition method based on HRRP has gradually become a research hotspot in the field of radar automatic target recognition. [0003] However, there are some problems in the traditional recognition methods, including: (1) Most of the feature extraction m...

AI Technical Summary

Problems solved by technology

[0003] However, there are some problems in the traditional recognition methods, including: (1) Most of the feature extraction methods are unsupervised and lossy, which means that the transformation-based feature extraction methods cannot focus on finding the maximum possible In terms of separable features, part of the separable information will inevitably be lost in the process of feature extraction, which is not conducive to the recognition of the back-end classifier

(2) The selection of feature extraction methods is highly dependent on the researchers' knowledge and experience of HRRP data, and it is difficult to achieve satisfactory results in the absence of prior information.

(3) Deep learning method based on recurrent neural network (RNN): This method is based on serial correlation modeling. Although the physical structure features are modeled and described, there are the following problems: (1) Time domain features of HRRP It is the most widely used in radar target classification, but the HRRP complex echo has a great relationship with the target attitude. Even if there is a slight change in the target attitude, the scattering points at the edge of the distance unit may move to several adjacent units. , which will have a great impact on the time-domain features of HRRP; (2) The original time-domain segmentation method is used for the local strength information of HRRP, and the obtained features are highly redundant, which brings great impact to the subsequent RNN modeling. Difficult; (3) One-way RNN can only use the structural information at the current moment and before the current moment when predicting, and cannot make good use of the overall structural information a priori contained in HRRP

Images