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

Supervision multi-view feature selection method based on automatic generation of view and unit with l1 and l2 norm minimization

A feature selection method and feature selection technology, applied in character and pattern recognition, instruments, computer components, etc., can solve the problem of information loss in high-resolution remote sensing images

Active Publication Date: 2015-11-11
HARBIN INST OF TECH
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to solve the problem of information loss in the feature selection process of high-resolution remote sensing images, and provides a method based on automatic view generation and joint l 1,2 Supervised multi-view feature selection method with norm minimization

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
  • Supervision multi-view feature selection method based on automatic generation of view and unit with l1 and l2 norm minimization
  • Supervision multi-view feature selection method based on automatic generation of view and unit with l1 and l2 norm minimization
  • Supervision multi-view feature selection method based on automatic generation of view and unit with l1 and l2 norm minimization

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0065] Specific implementation mode one: the following combination figure 1 and figure 2 Describe this embodiment, which is based on automatic view generation and combination1 1,2 A supervised multi-view feature selection method for norm minimization, which includes the following steps:

[0066] Step 1: collect remote sensing image data, preprocess the remote sensing image data; then perform feature extraction to obtain a feature vector set; then normalize all feature vectors in the feature vector set to obtain the original feature vector set;

[0067] Step 2: Using the affinity propagation algorithm to generate feature multi-views from the original feature vector set obtained in step 1;

[0068] Step 3: Based on l 1,2 Norm performs supervised multi-view feature selection on the feature multi-view generated in step two.

[0069] This implementation is divided into three steps, one is data preparation; the other is view generation; the third is based on l 1,2 Feature sele...

specific Embodiment approach 2

[0070] Specific implementation mode two: the following combination figure 1 and figure 2 Describe this embodiment. This embodiment will further explain Embodiment 1. The preprocessing of remote sensing image data in step 1 includes sequentially performing geometric fine correction and image registration, image mosaic and cropping, atmospheric correction and Bad tape removal.

[0071] This embodiment mainly includes bad band removal and data deformation and reorganization. Preprocessing is a very important step in remote sensing applications.

specific Embodiment approach 3

[0072] Specific implementation mode three: the following combination figure 1 and figure 2 Describe this embodiment mode. This embodiment mode will further explain Embodiment 1 or 2. The specific method for obtaining the feature vector set described in step 1 is: extracting the feature values ​​of the preprocessed remote sensing image data to obtain feature data, and all feature data From n samples x in the m-dimensional feature space i Composition, characteristic data recorded as which sample the y i is x i mark, y i ∈{1,...,c}, c is the number of categories; put n samples x i The row vector of is denoted as the sample set X: The label vector corresponding to the sample set X is y, and the n samples x i The m eigenvectors formed by the column vectors are denoted as X=[f 1 ,f 2 ,...,f m ],and

[0073] The method to obtain the original feature vector set in step 1 is: the feature data All the eigenvalues ​​in are mapped to [0-1] to obtain the original set of...

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

A supervision multi-view feature selection method based on automatic generation of view and unit with l1 and l2 norm minimization belongs to the technical field of remote sensing image data processing. The method provided by the invention is to solve a problem of information loss of a high-resolution remote sensing image in the feature selection process. The method comprises three steps: 1. acquiring remote sensing image data and pre-processing the same; then carrying out feature extraction to obtain an feature vector set; and normalizing all feature vectors of the feature vector set to obtain an original feature vector set; 2. generating multiple views from the original feature vector set obtained in step 1 by adopting an affinity propagation algorithm; and and 3. carrying out supervision multi-view feature selection on the multiple views obtained in step 2 based on the l1 and l2 norms. The invention relates to the supervision multi-view feature selection method.

Description

technical field [0001] The present invention relates to based on automatic view generation and federation l 1,2 A supervised multi-view feature selection method with norm minimization belongs to the technical field of remote sensing image data processing. Background technique [0002] With the development of imaging technology, remote sensing images are more and more widely used, such as the survey and detection of disasters and the environment, the update of basic geographic data and many other fields. There are many features that can be extracted from these remote sensing images, and high-dimensional features bring two challenges to the application of remote sensing images: the first is that high-dimensional features will lead to the curse of dimensionality. This problem can be solved with feature selection methods, which are also receiving increasing attention. Random forests combine feature selection and classification and can be used for landslide mapping. Locally we...

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
IPC IPC(8): G06K9/66
CPCG06V30/194
Inventor 陈曦张钧萍张晔
Owner HARBIN INST OF TECH
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