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

Based on approximate l 0 Remote sensing image unmixing method and system based on modified deep belief network

A deep belief network and remote sensing image technology, applied in image enhancement, image data processing, neural learning methods, etc., can solve the problems of parameter sensitivity and troublesome adjustment of multiple parameters, and achieve the effect of improving unmixing accuracy

Inactive Publication Date: 2021-03-30
CHINA UNIV OF GEOSCIENCES (WUHAN)
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Most of the traditional image unmixing methods use the linear unmixing model. In order to improve the accuracy of the unmixed abundance matrix, it is often necessary to add an image like L to the abundance matrix. 1 Norm and other constraints, although under many constraints, the accuracy of image unmixing is relatively improved, but the adjustment of multiple parameters is very troublesome, and the effect of unmixing is more sensitive to parameters

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
  • Based on approximate l <sub>0</sub> Remote sensing image unmixing method and system based on modified deep belief network
  • Based on approximate l <sub>0</sub> Remote sensing image unmixing method and system based on modified deep belief network
  • Based on approximate l <sub>0</sub> Remote sensing image unmixing method and system based on modified deep belief network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific implementation manners of the present invention will now be described in detail with reference to the accompanying drawings.

[0031] refer to figure 1 , which is a flow chart of the implementation of the remote sensing image sparse unmixing method proposed by the present invention, specifically including the following steps:

[0032] L1. Construct a set of data using a spectral library with a band of 224 and a feature type of 10, and divide the data into training data and test data in proportion;

[0033] L2. The training data is input to the deep belief network for network training; wherein, the specific steps of network training are:

[0034] A1. Since each feature point cannot contain all types of features, the output unmixed abundance matrix e(k) is actually sparse; based on the above sparse characteristics, the unmixed abundance matrix e...

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 deep belief network remote sensing image unmixing method and system based on approximate L0 transformation. A plurality of constraint conditions are added in a traditional image unmixing method, Although the accuracy of image unmixing can be improved, the result of the unmixing is very sensitive to the parameters;; Aiming at a traditional image unmixing method, the deep belief network is applied to remote sensing image unmixing, and meanwhile, the constraint condition of adding an approximate L0 norm is considered, so that the unmixing mode is simpler and more convenient while the unmixing precision is improved.

Description

technical field [0001] The present invention relates to the field of images, more specifically, to a method based on approximate L 0 Remote sensing image unmixing method and system based on modified deep belief network. Background technique [0002] Remote sensing images refer to films or photos that record the size of electromagnetic waves of various ground objects, and are mainly divided into aerial photos and satellite photos; remote sensing images themselves have a large amount of information and relatively high resolution, which makes their classification of ground objects To a certain extent, the ability has been greatly increased compared with the past. [0003] Most of the traditional image unmixing methods use the linear unmixing model. In order to improve the accuracy of the unmixed abundance matrix, it is often necessary to add an image like L to the abundance matrix. 1 Norm and other constraints, although under many constraints, the accuracy of image unmixing i...

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): G06T5/00G06N3/08
Inventor 李杏梅王心宇刘晓杰
Owner CHINA UNIV OF GEOSCIENCES (WUHAN)
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