Hyper-spectral image wave band dimension descending method based on NNIA evolutionary algorithm

An evolutionary algorithm, hyperspectral remote sensing technology, applied in remote sensing image processing, image classification, target recognition, mixed pixel decomposition fields, to achieve the effect of good physical characteristics

Active Publication Date: 2015-01-28
XIDIAN UNIV
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, unlike single-objective optimization problems, there is only one optimal solution. For multi-objective problems, one solution is better for one goal, and may be poor for other goals.

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
  • Hyper-spectral image wave band dimension descending method based on NNIA evolutionary algorithm
  • Hyper-spectral image wave band dimension descending method based on NNIA evolutionary algorithm
  • Hyper-spectral image wave band dimension descending method based on NNIA evolutionary algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] refer to figure 1 , the specific implementation steps of the present invention are as follows:

[0031] Step 1. Given the operating parameters, set the termination condition of the algorithm.

[0032] The operating parameters include: active antibody population size NA, clone size CS, non-dominated population size NM, evolutionary algebra gmax and input selection band number Num.

[0033] Preferably, NA is set to 20, CS is set to 100, NM is set to 100, and gmax is set to 100. Num is the number of bands required by the user, and the user needs to reduce to the number of dimensions and just input the number, which is set to 16 in the present invention.

[0034] Step 2. Input the original remote sensing image data, and convert the original data into L*M format, where L is the number of pixels in a band, and M is the number of bands of the original data. The real-number encoding method is adopted, and the band number represents this band, and the initial band combination...

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 hyper-spectral remote sensing image wave band dimension descending method based on an NNIA evolutionary algorithm. The problem that an existing hyper-spectral remote sensing image wave band dimension descending method changes physical significance of an original image and cannot completely retain interested information and remove redundant information is mainly solved. The hyper-spectral remote sensing image wave band dimension descending method comprises the steps of defining two objective functions to respectively represent the kurtosis of selected wave band combined information and KL divergence of wave band correlation, then utilizing an NNIA optimized objective function to obtain a group of non-dominated solutions and enabling a user to optionally select required results. The hyper-spectral remote sensing image wave band dimension descending method can be used in the technical fields of hyper-spectral remote sensing image classification, objective recognition, mixed pixel decomposition and the like and has the advantages that the physical significance of an original wave band is not changed, complete information can be well retained, the redundant information is removed, and classification accuracy is high after dimension descending.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, relates to search and selection, and can be used in technical fields such as image classification, target recognition, and mixed pixel decomposition. Background technique [0002] Band dimensionality reduction is an important step in hyperspectral remote sensing image processing. Hyperspectral image data contains narrow spectra of hundreds of continuous wavelengths. Its rich spectral information provides the potential for accurate object identification. However, its huge amount of data has brought serious problems in terms of data transmission, calculation and storage. Especially when the dimensionality of the data is high, there is a strong correlation between some bands, which contains a lot of redundant information, which makes the classification accuracy decrease with the increase of the dimensionality of the limited training samples, namely Hughes phenomenon appear...

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): G06T3/00G06N3/00
CPCG06T3/0031
Inventor 公茂果马晶晶镡永强马文萍张明阳刘嘉王驰
Owner XIDIAN UNIV
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