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

A Nonlinear Unmixing Method for Hyperspectral Imagery Based on Bilinear Mixture Model

A hyperspectral image and hybrid model technology, applied in character and pattern recognition, instrument, scene recognition, etc., can solve the problems of over-fitting results, local minima, complex calculations, etc., achieve good robustness, and reduce algorithm complexity The effect of reducing the calculation time

Active Publication Date: 2020-05-12
FUDAN UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, they have the problem that the calculation is too complex and often fall into local minima, and the collinear effect produced by the strong correlation between endmembers and virtual endmembers will also lead to overfitting of the results and be more sensitive to noise [9]

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
  • A Nonlinear Unmixing Method for Hyperspectral Imagery Based on Bilinear Mixture Model
  • A Nonlinear Unmixing Method for Hyperspectral Imagery Based on Bilinear Mixture Model
  • A Nonlinear Unmixing Method for Hyperspectral Imagery Based on Bilinear Mixture Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0066] In the following, simulation data and actual remote sensing image data are used as examples to illustrate the specific implementation of the present invention.

[0067] The hyperspectral remote sensing image nonlinear unmixing algorithm based on the bilinear mixing model used in the present invention is represented by GEAB-FCLS.

[0068] 1. Simulation data experiment

[0069] In this section, the GAEB-FCLS algorithm and the linear abundance estimation algorithm FCLS [10], the data-driven nonlinear unmixing method KFCLS [12] based on Gaussian kernel (the kernel parameters are obtained through the cross-validation method between 0.01-300) and The traditional solving algorithms corresponding to the three models of FM, GBM and PPNM: Fan-FCLS[6], GBM-GDA[7] and PPNM-GDA[8] for performance comparison. And use the root mean square error of abundance RMSE (Root Mean Square Error) and image reconstruction error RE (Reconstructed Error) two factors to evaluate the accuracy of the abund...

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 belongs to the technical field of remote sensing image processing, in particular to a hyperspectral image nonlinear unmixing method based on a bilinear mixed model. According to the geometric characteristics of the bilinear mixed model, the present invention converts the complex nonlinear unmixing problem into a simple linear unmixing problem by expressing the nonlinear mixed component in the data as a linear contribution that combines common nonlinear effect endpoints , and then combined with the linear unmixing algorithm to iteratively estimate the correct abundance. The invention starts from the mixed model of the hyperspectral observation pixel and combines its geometric and physical meanings, not only can effectively make up for the lack of linear unmixing, but also can better overcome the adverse effects brought by collinear effects. It has good robustness to noise and the number of endmembers, and can be used as an effective means to solve the nonlinear unmixing of hyperspectral remote sensing images. It has important application value in the high-precision unmixing of hyperspectral remote sensing images and the detection and recognition of ground targets.

Description

Technical field [0001] The invention belongs to the technical field of remote sensing image processing, and specifically relates to a method for nonlinear unmixing of hyperspectral images. Background technique [0002] Remote sensing technology is a new comprehensive technology developed in the 1960s. It is closely related to science and technology such as space, electron optics, computers, and geography. It is one of the most powerful technical methods for studying the earth's resources and environment. Hyperspectral remote sensing is a multi-dimensional information acquisition technology that combines imaging technology with spectral technology. The image has the characteristics of high spectral resolution and integrated map, which provides extremely rich information for the extraction and analysis of ground feature information. However, due to the low spatial resolution and the complex and diversity of the distribution of ground features, the pixels in the image are mostly mi...

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): G06K9/00
CPCG06V20/194G06V20/13
Inventor 杨斌王斌
Owner FUDAN 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