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

Hyperspectral image classification method based on F-EEMD (fast ensemble empirical mode decomposition)

A technology that integrates empirical modalities and hyperspectral images, applied in the field of remote sensing, can solve the problems of hyperspectral image preprocessing, such as long preprocessing time, noise addition, large decomposition times, and large computational load, so as to maintain superior performance, improve accuracy and classify Speed, the effect of reducing the number of support vectors

Inactive Publication Date: 2015-02-18
HARBIN HANGKONG BOCHUANG SCI & TECH
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the large number of IMF screening times K and the noise addition and decomposition times N of the ensemble empirical mode decomposition, this leads to a large amount of computation for the ensemble empirical mode decomposition, which leads to a long preprocessing time of the hyperspectral image

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
  • Hyperspectral image classification method based on F-EEMD (fast ensemble empirical mode decomposition)
  • Hyperspectral image classification method based on F-EEMD (fast ensemble empirical mode decomposition)
  • Hyperspectral image classification method based on F-EEMD (fast ensemble empirical mode decomposition)

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0043] Specific implementation mode one: as figure 1 As shown, the specific steps of the hyperspectral image classification algorithm method based on fast ensemble empirical mode decomposition in this embodiment are as follows:

[0044] Step 1: Analyze the influence of IMF screening times K and noise addition and decomposition times N on ensemble empirical mode decomposition.

[0045] This step first graphically analyzes the influence of different IMF screening times K and different noise adding and decomposition times N on the empirical mode decomposition results of the hyperspectral image set. Then, mutual information is used to analyze the influence of IMF screening times N and EEMD noise addition and decomposition times N on EEMD results.

[0046] Step 2: Propose a fast ensemble empirical mode decomposition algorithm (F-EEMD) based on the mutual information threshold, and use the fast ensemble empirical mode decomposition algorithm to perform feature extraction and featur...

specific Embodiment approach 2

[0064] Specific embodiment two: set forth the specific embodiment of the present invention below by the classification example of the hyperspectral data 92AV3C that AVIRIS sensor gathers in Northwestern Indiana, U.S.A.:

[0065] Execution step 1: The influence of the number of IMF screening K and the number of noise addition and decomposition N on the set empirical mode decomposition.

[0066] The influence of IMF screening times K and noise addition and decomposition times N on the ensemble empirical mode decomposition is as follows:

[0067] 1). The influence of IMF screening times K on the results of ensemble empirical mode decomposition: the spatial information of the 20th band is selected as the research object. Carry out ensemble empirical mode decomposition under different IMF screening times K for spatial information, such as Figure 3-10 shown. Among them, H3(k) represents IMF3 when the number of screening is k. It can be seen from the image changes that the intern...

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 hyperspectral image classification method based on F-EEMD (fast ensemble empirical mode decomposition), which relates to a hyperspectral image classification method in the remote sensing field. The hyperspectral image classification method disclosed by the invention comprises the following steps of: analyzing influence on an ensemble empirical mode decomposition result by an intrinsic mode function screening time K and a noise adding and decomposing time N; on the basis of an analysis result, proposing an F-EEMD algorithm based on a mutual information threshold value; carrying out characteristic extraction and characteristic selection on hyperspectral image waves by F-EEMD; and classifying a selected hyperspectral image characteristic composition by an SVM (support vector machine) classifier to obtain a classification result. According to the method disclosed by the invention, the precision and the classification speed of the hyperspectral image can be effectively improved, a support vector number required in the classification process can be effectively reduced, the pre-processing time of each wave band of the hyperspectral image is obviously less than that of a 2D-EEMD-SVM algorithm, and meanwhile, the excellent performance of the 2D-EEMD-SVM algorithm is favorably kept.

Description

technical field [0001] The invention relates to a hyperspectral image classification method in the field of remote sensing, in particular to a hyperspectral image classification method based on fast ensemble empirical mode decomposition. Background technique [0002] Hyperspectral remote sensing images have high spectral resolution and can provide almost continuous spectral curves of surface features for each pixel, so hyperspectral remote sensing can retrieve land details. At present, hyperspectral images have been widely used. Since the imaging spectrometer is easily affected by atmospheric molecular scattering and absorption, aerosol scattering and absorption, surface scattering, terrain, etc. during the transmission of radiant energy, the spectral shape of the hyperspectral data will be distorted, thereby introducing various noises. Therefore, it is necessary to study the feature extraction method of hyperspectral data. [0003] Empirical Mode Decomposition (EMD) is an...

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): G06K9/62
Inventor 沈毅王艳金晶
Owner HARBIN HANGKONG BOCHUANG SCI & 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