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

Hyper spectral image classification method based on wavelet threshold denoising and empirical mode decomposition

An empirical mode decomposition and hyperspectral image technology, applied in the field of remote sensing, can solve the problems of insufficient classification accuracy and large wavelet noise, and achieve the effect of fast classification, good separability, and improved accuracy.

Active Publication Date: 2012-05-02
HARBIN INST OF TECH
View PDF2 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The present invention is to solve the inapplicability of wavelet to nonlinear and non-stationary hyperspectral images in the traditional method and the large noise in the first intrinsic mode function decomposed by the empirical mode decomposition method, and the accuracy of the obtained classification results is not large enough In order to solve the problem, the empirical mode decomposition method is introduced on the basis of wavelet filtering, and the essential characteristics of the signal after wavelet filtering are adaptively extracted, and a hyperspectral image classification method based on wavelet threshold noise reduction and empirical mode decomposition is proposed.

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 classification method based on wavelet threshold denoising and empirical mode decomposition
  • Hyper spectral image classification method based on wavelet threshold denoising and empirical mode decomposition
  • Hyper spectral image classification method based on wavelet threshold denoising and empirical mode decomposition

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0025] Specific implementation mode 1. Combination figure 1 Describe this embodiment, a hyperspectral image classification method based on wavelet threshold noise reduction and empirical mode decomposition, the specific method is:

[0026] Step 1, two-dimensional wavelet threshold denoising of the hyperspectral image;

[0027] Step 2, performing empirical mode decomposition and image reconstruction on the hyperspectral image after wavelet noise reduction;

[0028] Step 3: Use the SVM classifier to classify the hyperspectral reconstructed image nIMFs to obtain classification accuracy.

specific Embodiment approach 2

[0029] Specific embodiment two, combine figure 2 This embodiment is described. The difference between this embodiment and the first embodiment is that the specific method of the two-dimensional wavelet threshold denoising of the hyperspectral image in the first step is:

[0030] Step 1.1. Input the hyperspectral data, and normalize the hyperspectral data to obtain the grayscale image x; select the wavelet decomposition layer number i=1, and perform the original signal x i Perform wavelet decomposition to obtain wavelet coefficient d, namely:

[0031] d=DWT{x}

[0032] Step 1.2. Calculate x i The wavelet threshold value of , and threshold the wavelet coefficients, namely:

[0033] d' = η τ (d)

[0034] Among them, η τ () is a threshold processing function with a threshold value of τ, using soft threshold denoising, as shown in the following formula:

[0035] d ′ = η τ ( ...

specific Embodiment approach 3

[0039] Specific embodiment three, combine image 3 , Figure 5 , Figure 6 , Figure 7 , Figure 8 , Figure 9 , Figure 10 , Figure 11 , Figure 12 , Figure 13 , Figure 14 and Figure 15 Describe this embodiment, the difference between this embodiment and specific embodiment 1 is that in step 2, the specific method for empirical mode decomposition and image reconstruction of the hyperspectral image after wavelet noise reduction is:

[0040] Step 2.1. Perform empirical mode decomposition on the hyperspectral image after wavelet noise reduction, and decompose it into q internal solid mode functions IMF and a residual:

[0041] Suppose X' k is the kth band image of the filtered hyperspectral image, x′(m,n) is the image X′ k Values ​​at pixel points (m, n) (m=1, 2,..., M; n=1, 2,..., N); r ij (m, n) is used to calculate the i (i=1, 2, ..., I) IMF process, the input value of the j (j = 1, 2, ..., J) iteration, the kth Band image X' k As an input signal, the origina...

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 hyper spectral image classification method based on wavelet threshold denoising and empirical mode decomposition relates to a hyper spectral image classification method in a remote sensing field. In the traditional method, the wavelet is not suitable for a non-linear and non-stationarity hyper spectral image, a first intrinsic mode function decomposed by using an empirical mode decomposition method has a loud noise and acquired classification result precision is not high. By using the method of the invention, the above problems can be solved. The method comprises the following steps: step 1. performing two-dimensional wavelet threshold denoising of the hyper spectral image; step 2. carrying out empirical mode decomposition and image reconstruction to the hyper spectral image which is performed with wavelet denoising; step 3. using a SVW classifier to classify the hyper spectral reconstruction image nIMFs so as to obtain the classification precision. The method can be used in the classification of the hyper spectral image.

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 wavelet threshold noise reduction and 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 are widely used in various fields such as agriculture, forestry, geological exploration, atmospheric monitoring, and military operations. Because the hyperspectral image is affected by many complex factors in the electromagnetic radiation propagation path of the sun-atmosphere-ground object-atmosphere-sensor and in the process of data transmission, various noises are introduced, which affects the classification accuracy of the hyperspectral image. ...

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): G06K9/62
Inventor 沈毅张敏张淼
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