SAR image target recognition method and device based on sparse representation and cascade dictionary

A sparse representation and target recognition technology, applied in the field of image processing and pattern recognition, can solve problems affecting the speed of solving sparse coefficients, affecting the speed of test sample recognition, high dictionary dimension, etc., to overcome the slow speed of sparse solving and good classification The effect of recognizing the effect

Pending Publication Date: 2020-11-27
NANJING COLLEGE OF INFORMATION TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When designing an over-complete dictionary, if the over-complete dictionary is constructed directly from the pixels of the SAR image or the features of the extracted training samples, the dictionary will have a high dimension and a high degree of redundancy, which will directly affect the subsequent sparse coefficient solving speed, thereby affecting Speed ​​of test sample recognition

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 target recognition method and device based on sparse representation and cascade dictionary
  • SAR image target recognition method and device based on sparse representation and cascade dictionary
  • SAR image target recognition method and device based on sparse representation and cascade dictionary

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0021] Embodiment 1, SAR image target recognition method based on sparse representation and cascaded dictionary, the flow chart is as follows figure 1 shown, including:

[0022] Segment the central area containing the target from the SAR image, remove the background noise to obtain the image to be recognized, and extract the single-shot amplitude, single-shot phase and single-shot azimuth features of the image to be recognized;

[0023] Based on the extracted monomorphic amplitude, monomorphic phase, and monomorphic azimuth features and the sub-dictionaries generated in advance for the monomorphic amplitude, monochromatic phase, and monochromatic azimuth features of the training sample images, the sparse coefficient is calculated by minimizing the L1 norm, The classification mechanism with the largest coefficient energy and the smallest reconstruction error is used for target classification recognition to obtain recognition results.

[0024] In this embodiment, optionally, a ...

Embodiment 2

[0078] Embodiment 2. On the basis of Embodiment 1, this embodiment provides a SAR image target recognition method based on sparse representation and cascaded dictionary. The method also includes:

[0079] Utilize the EMACH filter training sample image, generate the template image of the training sample image by every set azimuth (set every 12 degrees in this embodiment), specifically include;

[0080] (3.1) Input N training sample images, from left to right, from top to bottom to expand each pixel into a one-dimensional vector x i , where i=1,2,…,N, calculate x i the mean of the vector m;

[0081] (3.2) Define h as EMACH filter, FFT() means Fourier operation, let β∈(0,1), M=FFT(m), X i =FFT(x i ), calculate the intermediate parameter and as follows:

[0082]

[0083]

[0084] Among them, the symbol "+" represents matrix transposition;

[0085] (3.3) when the formula When the value is the largest, h is The eigenvectors corresponding to the eigenvalues ​​of th...

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 target recognition method based on sparse representation and a cascade dictionary. The method comprises the steps: firstly carrying out the template training of asample image based on a maximum extended average correlation height filter; secondly, extracting monogenic features of the template image, namely monogenic amplitude representing signal energy, monogenic phase representing signal structure information and monogenic orientation representing signal geometric information, constructing sub-dictionaries by the three features with complementary properties, and cascading the plurality of sub-dictionaries by using each sub-dictionary as a classifier; and finally, SAR image target classification is realized based on a classification mechanism with themaximum sparse representation coefficient energy and the minimum reconstruction error, and a good classification and recognition effect can be realized.

Description

technical field [0001] The invention relates to the field of image processing and pattern recognition, in particular to a SAR image target recognition method based on sparse representation and cascaded dictionary in the field of Synthetic Aperture Radar (SAR) image target recognition. Background technique [0002] In military battlefield surveillance and civilian real-time monitoring occasions, it is often necessary to classify or identify targets. SAR image target classification means that the radar detects the target, processes the echo information reflected by the target, and determines the attribute, category or type of the target. Because of the high-dimensional variability of target features, the complex background during imaging, and the variable factors of the SAR sensor itself, the classification and recognition of SAR images has become a difficult problem. In the process of acquiring SAR images, even if two identical targets belong to the same category, the differ...

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): G06K9/00G06K9/62G06K9/46G06K9/32G06K9/40
CPCG06V20/13G06V10/30G06V10/25G06V10/40G06F18/24
Inventor 季秀霞王肖
Owner NANJING COLLEGE OF INFORMATION 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

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