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

A Remote Sensing Quantitative Inversion Method for Sampling Learning Machine Adapted to Noise Conditions

A learning machine and inversion technology, applied in the field of remote sensing applications, can solve the problems of nonlinear noise and interference in the quantitative inversion of remote sensing, and achieve the effect of strong generalization ability, eliminating the influence of noise, and improving accuracy.

Inactive Publication Date: 2017-12-29
广西中马园区数字城市科技有限公司
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved by the present invention is to provide a more adaptable remote sensing quantitative inversion method, and at the same time solve the problems of nonlinearity and noise interference in the application of remote sensing quantitative inversion

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 Remote Sensing Quantitative Inversion Method for Sampling Learning Machine Adapted to Noise Conditions
  • A Remote Sensing Quantitative Inversion Method for Sampling Learning Machine Adapted to Noise Conditions
  • A Remote Sensing Quantitative Inversion Method for Sampling Learning Machine Adapted to Noise Conditions

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] As a nonlinear and statistical modeling tool, artificial neural network (ANN) is widely used in remote sensing quantitative inversion. The model can fully approximate any complex nonlinear relationship through the setting of neural structure and connection weights. However, due to the defects of the ANN model itself, such as slow learning convergence, easy to fall into local extremum, and difficult to determine the network structure, the inversion accuracy is difficult to meet the application requirements. Huang et al. improved the traditional ANN model and proposed an extreme learning machine (Extreme Learning Machine, ELM) algorithm. The ELM algorithm is a new type of single hidden layer feed-forward neural network, and its learning speed and generalization ability are greatly improved compared with the ANN model. This mainly depends on two aspects of improvement: (1) Randomly generate a small order of magnitude value as the connection weight between the input layer ...

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 present invention provides a remote sensing quantitative inversion method for sampling learning machines adapted to noise conditions, and uses the characteristics of fixed small order weights in extreme learning machines to simulate the nonlinear complex mathematics between influencing factors and inversion objects in remote sensing quantitative inversion Transform it into solving a linear system Hβ=TT; adaptively select the model parameter estimation algorithm according to the dimension of the network model parameter β; use the selected model parameter estimation algorithm to realize the solution of the network model parameter β in Hβ=TT. The present invention establishes a complex mathematical relationship model between influencing factors and inversion objects in remote sensing quantitative inversion; in the process of solving model parameters, it can filter the interference of sample data noise and adaptively select model parameter estimation algorithms, thereby quickly obtaining the model The best parameter results.

Description

technical field [0001] The invention relates to the field of remote sensing applications, in particular to a sampling learning machine remote sensing quantitative inversion method adapted to noise conditions. Background technique [0002] As an important means of earth system observation, remote sensing technology can provide continuous information on global land surface changes. In recent years, the application demand for quantitative inversion of water, atmospheric and ecological environment parameters based on remote sensing data has become increasingly prominent, and increasingly urgent requirements have been put forward for the accuracy of quantitative inversion. The main problem to be solved in quantitative remote sensing is how to use remote sensing data to accurately estimate surface parameters, realize the link of remote sensing data industry application models, and improve the prediction accuracy of the model. Taking the application of water quality remote sensing...

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): G06F19/00
Inventor 沈永林艾烨霜
Owner 广西中马园区数字城市科技有限公司
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