Dual deep neural network-based radar range profile target identification method

A deep neural network and target recognition technology, which is applied in the field of radar one-dimensional range image target recognition based on double deep neural networks, can solve the problems of over-fitting of deep models and affect the recognition results, and achieve the effect of accelerating the convergence speed.

Active Publication Date: 2017-12-12
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF7 Cites 35 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in target recognition based on one-dimensional range images, due to the small number of training samples, the d

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Dual deep neural network-based radar range profile target identification method
  • Dual deep neural network-based radar range profile target identification method
  • Dual deep neural network-based radar range profile target identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] The technical scheme of the present invention will be described in detail below in conjunction with examples.

[0043] First, the radar target backscattering simulation software is used to generate one-dimensional range image data of five types of aircraft targets, in which the sampling angle interval of each one-dimensional range image data is 0.1 degree attitude angle, and a total of 1800 one-dimensional image data are generated for each type of target , the dimension of each range image is 320 dimensions, and the original data set is: Among them, the j-th range image of the i-th type target is expressed as: For each one-dimensional range image, the distance perturbation is performed by randomly inserting 0 at the front and rear ends (the total number of inserted 0 is 80), so as to expand each range image into 10 images, and then add noise to the expanded data set , add 22db Gaussian white noise, and normalize the energy of the data set, record the processed data s...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention belongs to the technical field of radars and particularly relates to a dual deep neural network-based radar range profile target identification method. The method comprises the steps of firstly performing preprocessing operations of random distance disturbance, sample expansion, noise addition and the like on range profile data of a target to enhance the robustness of an identification system; secondly in combination with a deep learning theory, proposing a dual deep neural network (DDNN) with an adaptive learning rate, and performing unsupervised pre-training and supervised fine adjustment on the DDNN to obtain DDNN model parameters; thirdly performing pre-identification on test samples by utilizing the DDNN to obtain pre-identification results of the samples in two sub-networks; and finally according to the pre-identification results, performing time-space multi-level decision fusion by utilizing an improved DS evidence theory to obtain a target identification result.

Description

[0001] technical field [0002] The invention belongs to the technical field of radar, and in particular relates to a radar one-dimensional range image target recognition method based on a double deep neural network. Background technique [0003] With the continuous development of deep learning theory, related algorithms based on deep learning have been widely used in many target recognition fields. However, in target recognition based on one-dimensional range images, due to the small number of training samples, the deep model is prone to overfitting problems during the learning process, which ultimately affects the recognition results. Therefore, it is necessary to study a new deep learning model according to the characteristics of one-dimensional image data to further improve the recognition rate. Contents of the invention [0004] The purpose of the present invention is to provide a new dual-depth neural network model with adaptive learning rate for the target recognitio...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/62G06N3/02G01S7/41
CPCG06N3/02G01S7/417G06F18/254G06F18/29
Inventor 廖阔司进修周毅何旭东杨孟文周代英沈晓峰
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap