Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Multi-level-combined multi-look synthetic aperture radar image target recognition method

A synthetic aperture radar, target recognition technology, applied in image enhancement, image data processing, character and pattern recognition, etc., can solve problems such as target recognition rate limitation, and achieve the effect of improving the accuracy rate

Active Publication Date: 2014-07-30
菏泽建数智能科技有限公司
View PDF2 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, most of the methods for target recognition using multi-view SAR images under different azimuth angles of the same target are only carried out at one level of the target recognition framework, which limits the improvement of the target recognition rate.

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
  • Multi-level-combined multi-look synthetic aperture radar image target recognition method
  • Multi-level-combined multi-look synthetic aperture radar image target recognition method
  • Multi-level-combined multi-look synthetic aperture radar image target recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] The present invention will be further described below in conjunction with drawings and embodiments.

[0030] refer to figure 1 , a multi-level combined multi-view synthetic aperture radar image target recognition method, comprising the following steps:

[0031] Step 1, for the collected N pieces of unknown targets of the same kind under different azimuth angles, the multi-view synthetic aperture radar target image x k , k=1,2,...N for image preprocessing;

[0032] Step 2, performing feature extraction on each preprocessed SAR image using wavelet decomposition and principal component analysis, and extracting a multidimensional feature vector representing the image;

[0033] Step 3, classify the feature vector obtained in step 2 with a trained support vector machine that supports multi-objective classification, and obtain the image x k , k=1,2,...N belongs to the posterior probability p(q|x of a certain target category q k );

[0034] Step 4, for the preprocessed mul...

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

A multi-level-combined multi-look synthetic aperture radar image target recognition method comprises the following steps that firstly, multi-look synthetic aperture radar images under different azimuthal angles are preprocessed; secondly, feature extraction is carried out on the preprocessed images through wavelet decomposition and principal component analysis; thirdly, features are classified through a support vector machine, and the posterior probability that each image belongs to one class is obtained; fourthly, for the preprocessed images, a high-resolution radar image is rebuilt through a convex set projection super-resolution rebuilding algorithm at a data layer; fifthly, feature extraction is carried out on the rebuilt high-resolution image through wavelet decomposition and principal component analysis; sixthly, the features obtained from the fifth step are classified through the support vector machine, and the posterior probability that each rebuilt image belongs to one class is obtained; seventhly, decision-making layer fusion is carried out on the posterior probability of each single image and the posterior probability of the rebuilt image through a Bayesian decision fusion method with weights, and the classes of the multi-look synthetic aperture radar images are obtained.

Description

technical field [0001] The invention relates to the fields of image processing, pattern recognition and the like, in particular to the fields of synthetic aperture radar image processing and target recognition. Background technique [0002] Synthetic Aperture Radar (SAR) image target recognition is a key step in SAR image interpretation and processing, and has important application value. SAR image target recognition is to detect, locate and distinguish the type and model of the target on the acquired SAR image containing the target. SAR images are very sensitive to changes in target azimuth due to shadow effects, signal-background interaction, projection of 3D scenes on oblique planes, and other radar cross-section-sensitive factors. Therefore, the recognition ability of the target in the SAR image will also change with the change of the azimuth angle. Therefore, multi-view SAR images under different azimuth angles of the same target can be used to improve the performance...

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): G06K9/62G06T5/50
Inventor 宦若虹潘赟王楚郭峰
Owner 菏泽建数智能科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com