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

Classification Method of Polarized SAR Image Based on Deep Wavelet Neural Network

A technology of wavelet neural network and classification method, which is applied in the field of image processing, can solve the problems of unable to extract high-dimensional features and low classification accuracy, and achieve the effect of improving classification accuracy, strong approximation and fault tolerance, and high classification accuracy

Active Publication Date: 2017-07-28
XIDIAN UNIV
View PDF2 Cites 0 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 the S4VMs classifier, it cannot extract higher-dimensional features from the coherence matrix, resulting in low classification accuracy.

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
  • Classification Method of Polarized SAR Image Based on Deep Wavelet Neural Network
  • Classification Method of Polarized SAR Image Based on Deep Wavelet Neural Network
  • Classification Method of Polarized SAR Image Based on Deep Wavelet Neural Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] The present invention is a kind of classification method of polarimetric synthetic aperture radar SAR image based on depth wavelet neural network, referring to the attached figure 1 , describe in detail the specific implementation steps of the present invention:

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

[0037] Step 2: Preprocessing, use a 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, you can use 3×3, 5×5, 7 The coherence matrix of the polarimetric SAR image is filtered wit...

Embodiment 2

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

[0056]

[0057] Among them, ψ(j) represents the output of hidden layer node j, where ψ is a general representation of the output of hidden layer nodes, and the output of hidden layer nodes in the first layer network is represented by ψ 1 Indicates that the hidden layer node output of the second layer network is represented by 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 characteristic number of the coherence matrix, where the value is 9, and the input nodes of the second layer network are equal to the hidden layer nodes of the first layer network number, here the value is 100, W′ jk Indicates the weight between the hidden layer node j and...

Embodiment 3

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

[0064]

[0065] Among them, E represents the mean square error of the sample, where E is a general representation of the mean square error, and the error of the first layer network is represented by E 1 Indicates that the error of the first layer network is expressed by E 2 Indicates that S is the number of training samples, 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, here the value is 9, the number of output nodes of the second layer network is equal to the number of second layer The number of input nodes of the network, the value here is 100, h(i) represents the output of the output node i, where h is a general representation of the output laye...

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 polarimetric SAR image classification method based on a deep wavelet neural network, which mainly solves the problem in the prior art that the classification accuracy decreases due to less number of features or unreasonable feature extraction. The implementation steps are: input image; preprocessing; selecting samples; using training samples to train deep wavelet neural network; extracting features; classifying; calculating classification accuracy. The present invention trains the deep wavelet neural network in a layer-by-layer manner, which avoids the problem of gradient diffusion when the number of network layers is large, and can extract the essential characteristics of the data, describe the detailed characteristics of the data, and highlight the differences between different types of ground objects. Dimensional features. Since the present invention utilizes the deep wavelet neural network to extract the deep high-dimensional features of the data, it successfully avoids the problem of less feature numbers or insufficient feature learning in the classification technology, and improves the classification of polarimetric synthetic aperture radar SAR images. precision.

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
Patent Type & Authority Patents(China)
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 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