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

Nonlinear unmixing method for hyperspectral images based on kernel sparse nonnegative matrix factorization

A non-negative matrix decomposition, hyperspectral image technology, applied in the field of remote sensing information processing, can solve problems such as unconsidered, achieve the effect of solving nonlinear unmixing, overcoming linear unmixing, and good anti-noise performance

Inactive Publication Date: 2017-06-06
扬州匠新精密数控设备有限公司
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Because it is an unmixing method based on a linear model, the influence of the actual natural environment on hyperspectral unmixing is not considered. Due to the existence of nonlinear mixing phenomena, such methods will not get accurate results

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
  • Nonlinear unmixing method for hyperspectral images based on kernel sparse nonnegative matrix factorization
  • Nonlinear unmixing method for hyperspectral images based on kernel sparse nonnegative matrix factorization
  • Nonlinear unmixing method for hyperspectral images based on kernel sparse nonnegative matrix factorization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] Now in conjunction with embodiment, accompanying drawing, the present invention will be further described:

[0038] An embodiment of the present invention is a hyperspectral image nonlinear unmixing method based on kernel sparse non-negative matrix decomposition, and the specific steps are as follows:

[0039] Step 1 Input the hyperspectral data to be unmixed Where L is the number of bands and N is the number of pixels.

[0040] Step 2 uses the virtual dimension approach to The number of endmembers is estimated to be P in the hyperspectral image.

[0041] Step 3: Input the unmixing parameters: respectively input the number of endmembers P, the coefficients λ and β, the kernel parameters σ and the minimum convergence residual threshold τ.

[0042] Step 4 Construct the following kernel function matrix K

[0043]

[0044] Where k is the kernel function, represents the inner product, and the value of σ is 1.

[0045] Step 5 uses an alternate iterative optimizatio...

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 relates to a hyperspectral image nonlinear unmixing method based on kernel sparse non-negative matrix decomposition, which uses the virtual dimension method to estimate the number of end members of the hyperspectral image, and then uses the kernel method to combine the traditional linear mixed model based The unmixing algorithm is generalized to nonlinear feature spaces, and an alternate iterative optimization method is used to solve the nonlinear spectral unmixing problem. The beneficial effect is that it starts from the mixed model of hyperspectral observation pixels, adds the sparsity of hyperspectral abundance to the model, and then maps the linear mixed model to the nonlinear mixed model through the kernel method, effectively overcoming the linear unmixing At the same time, it has good anti-noise performance and can be used as an effective means to solve the nonlinear unmixing of hyperspectral remote sensing images.

Description

technical field [0001] The invention belongs to the technical field of remote sensing information processing, and in particular relates to a hyperspectral image nonlinear unmixing method based on kernel sparse non-negative matrix decomposition. Background technique [0002] Hyperspectral remote sensing images are widely used in remote sensing applications due to their advantages of high spectral resolution and multiple imaging bands. Due to the generally low spatial resolution and the complex diversity of natural objects, an observed pixel is likely to contain more than one type of object, and this pixel is called a mixed pixel. Mixed pixels commonly exist in hyperspectral remote sensing images. Therefore, the technology of analyzing the proportion of various types of ground objects in mixed pixels, that is, spectral unmixing technology, is the prerequisite for accurate classification and identification of ground objects. The mixed pixel decomposition methods of hyperspectr...

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): G06K9/62
CPCG06F18/29
Inventor 李映房蓓
Owner 扬州匠新精密数控设备有限公司
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