SAR image ship classification method fusing dual-polarization features

A classification method and dual-polarization technology, applied in the field of synthetic aperture radar image interpretation, can solve the problems of insufficient multi-resolution analysis of features and insufficient utilization of polarization features of SAR images, so as to enrich the characteristics of SAR ships and suppress speckle noise. , the effect of improving the accuracy

Pending Publication Date: 2022-01-28
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The invention belongs to the technical field of synthetic aperture radar (SAR) image interpretation, and the invention discloses a method for classifying ships in SAR images that combines dual-polarization features and uses Laplacian pyramids for multi-resolution analysis, and is used to solve Problems existing in the prior art include insufficient utilization of SAR image polarization features and insufficient multi-resolution analysis of features

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 image ship classification method fusing dual-polarization features
  • SAR image ship classification method fusing dual-polarization features
  • SAR image ship classification method fusing dual-polarization features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0110]Attached below figure 1 And attached figure 2 The present invention is further described in detail.

[0111] Step 1. Prepare the dataset

[0112] The classic OpenSARShip dataset is obtained from Definition 1, and the single-view complex domain data is selected in the classic OpenSARShip dataset.

[0113] Three types of ship data sets are established. The types of ships are bulk carriers, container ships and cruise ships, and the numbers are 333, 573 and 242 respectively. 70% of the minimum number of ships among the three types of ships (242*70%=169) is used as the training set, the training samples of the remaining types of ships are also set to 169, and the remaining samples are set as the test set, so as to obtain category balance The three types of ship datasets.

[0114] Establish six types of ship data sets, the ship types are bulk carriers, cargo ships, container ships, fishing boats, general cargo ships and cruise ships. The number of these six types of ship...

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 image ship classification method fusing dual-polarization features. The SAR image ship classification method aims at ship classification problems in SAR images and designs four modules based on a deep learning method, namely a dual polarization feature fusion module (DPFF), a compression excitation module, a Laplacian pyramid module and a Softmax classifier module. The dual-polarization feature fusion module makes full use of polarization features, enriches SAR ship features and suppresses speckle noise; the compression excitation module balances the contribution of each polarization feature; and the Laplacian pyramid module extracts SAR ship features from coarse resolution to fine resolution, so precision is improved. After training is conducted and input images are sequentially processed by the four modules, classification of ships in the SAR images can be finally achieved; and the SAR ship classification method provided by the invention is superior to an existing SAR ship classification method based on deep learning in the aspect of classification accuracy.

Description

technical field [0001] The invention belongs to the technical field of synthetic aperture radar (SAR) image interpretation, and relates to a method for classifying ships in SAR images that combines dual polarization features. Background technique [0002] Synthetic Aperture Radar (SAR) is an active microwave remote-sensing radar that can observe and image the sea and land environment from long distances and from multiple angles all day long. Compared with optical imaging technology, synthetic aperture radar imaging technology can avoid the influence of climate and time, and has the characteristics of high resolution and good timeliness. It is widely used in the national economy and military fields such as exploration and natural disaster monitoring. For details, please refer to the literature "Hou Xiaohan, Jin Guodong, Tan Lining. A review of ship target detection in SAR images based on deep learning [J]. Laser and Optoelectronics Progress, 2021,58(04):53-64.". [0003] Ma...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06V10/764G06V10/80G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/2415G06F18/254
Inventor 张晓玲王宝有张天文胥小我师君韦顺军
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