SAR (Synthetic Aperture Radar) three-dimensional rotating ship target refocusing method based on CV-RefocusNet

A three-dimensional rotation and refocusing technology, applied in neural learning methods, radio wave reflection/re-radiation, measurement devices, etc., can solve the problem of unrecognizable SAR ship target images, achieve good image processing effects, and facilitate implementation Effect

Active Publication Date: 2020-09-29
HARBIN INST OF TECH
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem in the prior art that blurred SAR ship target images cannot be recognized and obtained reliable re

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 (Synthetic Aperture Radar) three-dimensional rotating ship target refocusing method based on CV-RefocusNet
  • SAR (Synthetic Aperture Radar) three-dimensional rotating ship target refocusing method based on CV-RefocusNet
  • SAR (Synthetic Aperture Radar) three-dimensional rotating ship target refocusing method based on CV-RefocusNet

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0043] Specific implementation mode 1. Combination Figure 1 to Figure 4 Shown, the present invention provides a kind of SAR three-dimensional rotating ship target refocusing method based on CV-RefocusNet, comprising,

[0044]The SAR ship target image is obtained based on the 3D ship model and the ray tracing method simulation, and the SAR ship target image includes the SAR 3D rotating ship target image as a sample and the SAR stationary ship target image as a label; the SAR ship target image is The target image is grouped into a training sample library and a test sample library;

[0045] Construct complex domain convolutional neural network CV-RefocusNet framework, described CV-RefocusNet framework comprises an input layer, four convolution layers and four deconvolution layers, wherein the 4th deconvolution layer is as output layer; CV- The input, output, activation function and weight of the RefocusNet architecture all belong to the complex field;

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 discloses an SAR three-dimensional rotating ship target refocusing method based on CV-RefocusNet, and belongs to the field of SAR image processing. The method comprises three parts of SAR sample generation, network architecture design and refocusing implementation. The SAR sample generation part is responsible for generating an SAR three-dimensional rotating ship target image, a training sample library and a test sample library are formed, a network architecture design part gives a CV-RefousNet network architecture and clarifies network design details, and a refocusing implementation part performs training test on an SAR three-dimensional rotation ship target sample library based on CV-RefousNet to achieve a refocusing function. According to the invention, SAR three-dimensional rotation ship target refocusing can be realized, and a clear target image is obtained.

Description

technical field [0001] The invention relates to a CV-RefocusNet-based SAR three-dimensional rotating ship target refocusing method, belonging to the field of SAR image processing. Background technique [0002] The ship target sails on the sea, affected by the sea conditions and its own movement, there is a three-dimensional rotation. The imaging result of the same ship target is relatively clear in the static state, but in the three-dimensional rotating state, the imaging result is very blurred. This is because there is a three-dimensional rotation component of the ship target at sea, which makes the SAR (Synthetic Aperture Radar) imaging results blurred under high sea conditions, and the higher the sea conditions, the more serious the blur. The fuzzy SAR ship target image brings great difficulties to the subsequent information extraction and recognition process. [0003] Existing SAR information extraction and recognition algorithms can only achieve more accurate recognit...

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): G06K9/62G06N3/04G06N3/08G01S13/90G01S13/937
CPCG06N3/08G01S13/9004G01S13/937G06V2201/07G06N3/045G06F18/214
Inventor 张云化青龙姜义成徐丹
Owner HARBIN INST OF TECH
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