Hyperspectral image target detection method based on tensor matching subspace

A hyperspectral image and target detection technology, which is applied in image enhancement, image analysis, image data processing, etc., can solve the problems of low utilization rate of spatial information and low accuracy of target detection in hyperspectral images, and achieve the goal of improving the utilization rate of spatial information Effect

Active Publication Date: 2018-09-07
黑龙江省工研院资产经营管理有限公司
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to propose a hyperspectral image target detection method based on tensor matching subspace in order to solve the problems of low detection accuracy and low utilization rate of spatial information in the existing hyperspectral image

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
  • Hyperspectral image target detection method based on tensor matching subspace
  • Hyperspectral image target detection method based on tensor matching subspace
  • Hyperspectral image target detection method based on tensor matching subspace

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0023] Specific implementation mode one: combine figure 1 Describe this embodiment, the hyperspectral image target detection method based on tensor matching subspace in this embodiment, specifically referred to as:

[0024] Step 1. Establish the target sample H under the tensor representation 1 and the background sample H 0 The signal representation model of

[0025] Step 2. According to the hyperspectral sample data and window size, respectively establish the target sample H 1 and the background sample H 0 The space X, space Y, spectrum and fourth-order tensor matrix of atoms;

[0026] Step 3. According to the tensor matching subspace projection algorithm, obtain the target sample H 1 and the background sample H 0 Orthogonal projection matrices of the space X, space Y and spectrum three background directions and three target directions in the fourth-order tensor matrix of space X, space Y, spectrum and atom;

[0027] The data space of the signal to be detected under th...

specific Embodiment approach 2

[0030] Specific implementation mode two: combination Figure 2a , Figure 2b This embodiment is described. The difference between this embodiment and the first embodiment is that the target sample H under the tensor representation is established in the first step. 1 and the background sample H 0 The signal representation model of ; the specific process is:

[0031]

[0032] in, Represents the third-order tensor representation of the signal to be detected; Indicates the fourth-order tensor space formed by background samples; Indicates the fourth-order tensor quantum space formed by the target sample; x 4 Indicates that the weighted sum operation is performed on the fourth dimension of the tensor; α and β represent the corresponding abundance coefficients, that is, the corresponding weights; is a third-order tensor representation of Gaussian random noise;

[0033] Other steps and parameters are the same as those in Embodiment 1.

specific Embodiment approach 3

[0034] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is that in the step two, according to the hyperspectral sample data and the window size, the target sample H is respectively established 1 and the background sample H 0 The space X, space Y, spectrum and fourth-order tensor matrix of atoms; the specific process is called:

[0035] Step 21, randomly selecting hyperspectral sample data from the hyperspectral sample database and setting the window size of the hyperspectral sample data;

[0036] Step 22: Convert the background sample of each selected hyperspectral sample data into a third-order tensor form, and then form all the background samples in the third-order tensor form into a fourth-order tensor;

[0037]Step two and three, converting the target sample of each selected hyperspectral sample data into a form of third-order tensor, and then forming all target samples in the form of third-order tensor into a fourth-ord...

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 hyperspectral image target detection method based on tensor matched subspace, which relates to a hyperspectral image target detection method and aims at solving problems that the existing hyperspectral image target detection precision is low and the space information utilization rate is low. The method comprises specific steps: 1, signal representation models for a target and a background under tensor representation are built; 2, four-order tensor matrixes for the target and the background are built respectively; 3, an orthogonal projection matrix in three background directions and three target directions of space X, space Y and a spectrum in the four-order tensor matrixes for the target and the background is solved; 4, a to-be-detected signal is mapped to target sample projection subspace and background sample projection subspace obtained in the third step; and 5, whether the to-be-detected signal is a detection target is judged, if a generalized likelihood ratio detection model value under the tensor representation is larger than or equal to eta, the to-be-detected signal is the detection target, or otherwise, the to-be-detected signal is the background target, wherein eta is a threshold. The method of the invention is used in the image detection field.

Description

technical field [0001] The invention relates to a hyperspectral image target detection method. Background technique [0002] The hyperspectral sensor mainly extracts the reflection information of ground objects in different optical or near-infrared bands to obtain the map-spectral features of the ground. The hyperspectral image contains the spectral features of the ground objects and the spatial characteristics of the ground objects. This almost continuous Spectral sampling information can reflect small differences in the spectrum of ground objects, so this feature is often called a diagnostic feature, which is used as the basis for classification and detection of ground objects. It has important theoretical significance and application value to study the new technology of hyperspectral image target detection. In military terms, camouflage, concealment and deception of enemy targets can be revealed. In terms of civilian use, there have been important applications in public...

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 Patents(China)
IPC IPC(8): G06T7/33
CPCG06T2207/10036
Inventor 谷延锋刘永健高国明
Owner 黑龙江省工研院资产经营管理有限公司
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