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

Regularized high-order statistics based hyperspectral space multi-target detection method

A technology of high-order statistics and detection methods, which is applied in the field of hyperspectral remote sensing image target detection, and can solve the problem of not using high-order statistics of data.

Inactive Publication Date: 2011-08-17
BEIHANG UNIV
View PDF1 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] These existing target detection methods mainly use the second-order statistics of the data for calculation, mainly involving the covariance matrix or correlation matrix of the data, but do not use the high-order statistics of the data

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
  • Regularized high-order statistics based hyperspectral space multi-target detection method
  • Regularized high-order statistics based hyperspectral space multi-target detection method
  • Regularized high-order statistics based hyperspectral space multi-target detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0051] In order to better understand the technical solution of the present invention, the embodiments of the present invention will be further described below in conjunction with the accompanying drawings:

[0052] The present invention is realized under the MATLAB R2010a language environment. After the computer reads the hyperspectral remote sensing image data, it obtains a data cube. First, the data is de-averaged to make the mean value of the data zero, and then the data is whitened to remove the correlation of the data. The detection process can be regarded as a filtering process. Spectral curve x=[x 1 , x 2 ,...,x M ] T As the input of the filter, the filter weight vector w=[w 1 ,w 2 ,...,w M ] T and the product w of the input x T x as output. Set the high-order statistics of the output data as the objective function, find the optimal weight vector w, and add a negative regularization term to minimize the high-order statistics of the output data, and the selecti...

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 a regularized high-order statistics based hyperspectral space multi-target detection method, which comprises the four major steps of: 1, reading hyperspectral image data by a computer in an MATLAB (MATrix LABoratory) R2010a environment; 2, preprocessing the data by the computer, namely de-equalizing and whitening the data; 3, constructing a detection filter to minimally output high-order statistics of the data after a negative regularized item is added, and solving an optimal weight vector of the detection filter; and 4, setting a proper threshold to acquire a detection result image. Through the method, the defects of the prior art are overcome, the high-order statistics of the data is fully utilized, a good detection effect is achieved in spite of size of a target, and particularly, detection probability under a low false alarm rate condition can be increased. The method has high practical value and broad application prospect in the technical field of hyperspectral remote sensing image target detection.

Description

(1) Technical field: [0001] The invention relates to a hyperspectral space multi-target detection method based on regularized high-order statistics, and belongs to the technical field of hyperspectral remote sensing image target detection. (two) background technology: [0002] Today, there are more than 30,000 man-made objects in Earth orbit: all kinds of military and commercial satellites, rocket projectiles, and all kinds of debris and space junk. Some military powers are also researching various anti-satellite weapons. Space early warning has attracted more and more attention from various countries. More and more monitoring satellites are equipped with hyperspectral imaging systems, and the acquired hyperspectral images contain a lot of potential information. This is because the hyperspectral remote sensing system integrates the information of the image dimension and the spectrum dimension. When acquiring the image of the space target, the spectrum information is obtain...

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): G06T7/00G06T5/00
Inventor 史振威秦臻
Owner BEIHANG UNIV
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