Deep wavelet neural network-based polarimetric SAR (synthetic aperture radar) image classification method

A wavelet neural network and classification method technology, applied in the field of image processing, can solve the problems of inability to extract high-dimensional features and low classification accuracy, and achieve the effects of improving classification accuracy, strong approximation and fault tolerance, and good robustness.

Active Publication Date: 2015-03-11
XIDIAN UNIV
View PDF7 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that since this method directly uses the coherence matrix of polarimetric SAR images to train

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
  • Deep wavelet neural network-based polarimetric SAR (synthetic aperture radar) image classification method
  • Deep wavelet neural network-based polarimetric SAR (synthetic aperture radar) image classification method
  • Deep wavelet neural network-based polarimetric SAR (synthetic aperture radar) image classification method

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0034] Example 1:

[0035] The present invention is a classification method of polarization synthetic aperture radar SAR image based on deep wavelet neural network. figure 1 , A detailed description of the specific implementation steps of the present invention:

[0036] Step 1: The input image is actually the coherence matrix of a polarized synthetic aperture radar SAR image to be classified, see figure 2 , figure 2 Shown is the L-band multi-view polarization SAR image of Flevoland, Netherlands obtained by the AIRSAR platform in 1989. The coherence matrix of the image is a matrix of size 3×3×N, where N is the polarization synthetic aperture radar SAR image The total number of pixels.

[0037] Step 2: Preprocessing. Use Lee filter with a window size of 7×7 to filter the above coherence matrix to obtain the filtered coherence matrix. In specific simulation experiments, 3×3, 5×5, 7 ×7 equal-size windows filter the coherence matrix of the polarized synthetic aperture radar SAR image, ...

Example Embodiment

[0054] Example 2:

[0055] The polarization synthetic aperture radar SAR image classification method based on the deep wavelet neural network is the same as in embodiment 1, wherein the calculation hidden layer output formula described in step 4b and step 4e is as follows:

[0056] ψ ( j ) = exp ( - ( X k = 1 m W jk ′ x k - b j a j ) 2 / 2 ) cos ( 5 X ( X k = 1 m W jk ′ x k - b j a j ) )

[0057] Among them, ψ(j) represents the output of the hidden layer node j, where ψ is a general representation of the output of the hidden layer node, and the output of the hidden layer node of the first layer network is ψ 1 Indicates that the output of the hidden layer node of the second layer network is ψ 2 Indicates that m is the number of input nodes. In this example, the number of input nodes of the first layer network is equal to the number of fea...

Example Embodiment

[0062] Example 3:

[0063] The polarization synthetic aperture mine SAR image classification method based on the deep wavelet neural network is the same as in the embodiment 1-2, and the mean square error formulas described in step 4c and step 4f are as follows:

[0064] E = X s = 1 S X i = 1 n ( h ( i ) - x i ) 2

[0065] Among them, E represents the mean square error of the sample, where E is a total representation of the mean square error, and the error of the first layer network is E 1 Indicates that the error of the first layer network is E 2 Indicates that S is the number of training samples and n is the number of output nodes. In this example, the number of output nodes of the first-layer network is equal to the number of input nodes, and the value here is 9, and the number of output nodes of the second-layer network is equal to the second layer The number of input nodes of the network, here is the value 100, h(i)...

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 wavelet neural network-based polarimetric SAR (Synthetic Aperture Radar) image classification method, and aims to mainly solve the problem of classification accuracy reduction caused by fewer characteristics or unreasonable characteristic extraction in the prior art. The method is implemented by the following steps: inputting an image; performing preprocessing; selecting samples; training a deep wavelet neural network by utilizing a training sample; extracting characteristics; performing classification; calculating classification accuracy. According to the method, the deep wavelet neural network is trained layer by layer, so that the problem of gradient diffusion in case of more network layers is solved; moreover, high-dimensional characteristics capable of reflecting essential properties of data, describing detail characteristics of the data and highlighting differences between different ground object types can be extracted; the deep high-dimensional characteristics of the data are extracted by virtue of the deep wavelet neural network, so that the problem of fewer characteristics or incomplete and unreasonable characteristic learning in a classification technology is successfully solved, and the classification accuracy of a polarimetric SAR image is improved.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to the processing of polarimetric SAR images, in particular to a method for classifying polarimetric SAR images based on a deep wavelet neural network. It can be used to classify and identify ground targets in polarimetric SAR images. Background technique [0002] Synthetic aperture radar is a high-resolution imaging radar. Because microwaves have penetrating properties and are not affected by light intensity, synthetic aperture radars have all-day and all-weather working capabilities. Compared with other sensor images, it can show more details and can better distinguish the characteristics of nearby objects. With the development of technology, synthetic aperture mines are gradually developing in the direction of high resolution, multi-polarization, and multi-channel. Compared with traditional single-polarization SAR, multi-polarization SAR can provide richer target info...

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): G06K9/66G06N3/08
CPCG06N3/082G06F18/2411
Inventor 焦李成李玲玲姜红茹屈嵘杨淑媛侯彪王爽刘红英熊涛马文萍马晶晶
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap