SAR target classification method based on SAGAN sample expansion and auxiliary information

A technology for auxiliary information and target classification, which is applied in the field of small sample target recognition of synthetic aperture radar, and can solve the problems of small amount of SAR small sample data and so on.

Active Publication Date: 2019-06-25
HARBIN ENG UNIV
View PDF7 Cites 34 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The present invention mainly solves the problem of target areas in SAR remote sensing images. Based on the generation confrontation network, a network more suitable for SAR remote sensing images is propose...

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
  • SAR target classification method based on SAGAN sample expansion and auxiliary information
  • SAR target classification method based on SAGAN sample expansion and auxiliary information
  • SAR target classification method based on SAGAN sample expansion and auxiliary information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0056] The present invention will be further described below in conjunction with the accompanying drawings.

[0057] A SAR target classification method based on SAGAN sample expansion and auxiliary information, comprising the following steps:

[0058] 1. Input the noise z of noise(z) into the sample generator network with a self-attention mechanism to obtain a data image imitating real samples. It is processed through four similar modules L1, L2, L3, and L4 with different scales in turn. Each module contains convolution, spectral normalization, ReLU activation, and three data processing layers in turn. After passing through L3 and L4, each enters a scale Different self-attention mechanism layers, and then output image labels after passing through a convolutional layer and Tanh activation layer.

[0059] Deconvolution is to convolve the initial input small data (noise), and the size becomes larger.

[0060] The first is to invert the convolution kernel. Then the convolution ...

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 field of synthetic aperture radar small sample target recognition, and particularly relates to an SAR target classification method based on SAGAN sample expansion and auxiliary information. According to the characteristics of an SAR data sample image, the Inception structure is optimized and improved, proper regularization conditions are added, and the SAR small sampletarget is accurately identified by combining the GAN small sample generation and the GAN small sample super-resolution result. The invention provides a network more suitable for an SAR remote sensingimage, so that the network can learn the characteristics of different types of target areas, a new more realistic target area image is generated, and the problem of small data volume of an SAR smallsample is solved. A target area in a synthetic aperture radar SAR remote sensing image is aimed at. The invention discloses an SAR target classification method based on self-attention generative adversarial network sample expansion and auxiliary information, and mainly relates to a generative adversarial network for expanding SAR new sample data, and the method is used for SAR small sample targetidentification based on a Restnet50 structure network.

Description

technical field [0001] The invention belongs to the field of synthetic aperture radar small-sample target recognition, in particular to a SAR target classification method based on SAGAN sample expansion and auxiliary information. Background technique [0002] Synthetic aperture radar, or SAR, is an active microwave imaging sensor. By transmitting broadband signals and combining synthetic aperture technology, SAR can simultaneously obtain two-dimensional high-resolution images in both range and azimuth directions. Compared with traditional optical remote sensing and hyperspectral remote sensing, SAR has all-weather and all-weather imaging capabilities, and has a certain degree of penetration. The obtained images can reflect the microwave scattering characteristics of the target. technical means. SAR has been widely used in the fields of military and people's livelihood, and it is an important technical means to realize space military reconnaissance, natural resource census, ...

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): G06K9/62G06N3/04G06N3/08
Inventor 关键刘加贝孙建国王嘉岐吴嘉恒袁野
Owner HARBIN ENG UNIV
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