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

Polarized SAR Classification Method Based on Scattering Fully Convolutional Model

A convolution model and classification method technology, applied in the field of image processing, can solve problems such as low classification accuracy, incomplete scattering feature information, poor classification effect of image edge pixels, etc., to improve classification accuracy, improve training speed, The effect of speeding up training

Active Publication Date: 2021-09-03
XIDIAN UNIV
View PDF12 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 the scattering feature information extracted by this method is incomplete, so the classification effect on the edge pixels of the image is poor.
Although this method uses multi-target decomposition to obtain comprehensive polarization characteristics, the disadvantage of this method is that the method does not learn the texture characteristics of the image, resulting in incomplete feature information and low classification accuracy. The training speed of SVM is much slower than that of convolutional neural network

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
  • Polarized SAR Classification Method Based on Scattering Fully Convolutional Model
  • Polarized SAR Classification Method Based on Scattering Fully Convolutional Model
  • Polarized SAR Classification Method Based on Scattering Fully Convolutional Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] Attached below figure 1 The steps of the present invention are further described in detail.

[0030] Step 1. Perform Lee filtering on the polarimetric SAR image to be classified.

[0031] The scattering matrix of the polarimetric SAR image to be classified is subjected to refined polarization Lee filtering to filter coherent noise, and the filtered scattering matrix is ​​obtained. The size of the polarimetric SAR image to be classified is 1800×1380 pixels, and the filtered scattering matrix is ​​obtained. Each element in the matrix is ​​a 3×3 matrix, which is equivalent to a 9-dimensional feature for each pixel.

[0032] The window size of the Lee filter in the refined polarized Lee filter is 7×7 pixels.

[0033] Step 2. Perform Pauli decomposition of the scattering matrix.

[0034] The filtered scattering matrix is ​​decomposed by Pauli to obtain odd scattering energy, even scattering energy and volume scattering energy, and use the decomposition obtained odd scatte...

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 polarization SAR classification method based on a scattering full convolution model, the steps of which are: (1) performing Lee filtering on the polarization SAR image to be classified; (2) performing Pauli decomposition on the scattering matrix; (3) Normalization of the feature matrix; (4) Constructing a data set; (5) Constructing a scattering full convolution network model; (6) Training a scattering full convolution model; (7) Obtaining test results. The invention effectively combines the polarization characteristics, scattering characteristics and texture characteristics of the polarimetric SAR image, retains the integrity of the feature information, improves the classification accuracy of the image, and accelerates the training speed at the same time.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a polarization synthetic aperture radar (Synthetic Aperture Radar, SAR) image classification method based on a scattering (Scatter) full convolution model in the technical field of polarization synthetic aperture radar image ground object classification . The invention can be used to classify the ground objects of the polarimetric SAR images, can effectively improve the classification accuracy of the polarimetric SAR images, and can be used for the classification of the ground objects of the polarimetric SAR images and target recognition. Background technique [0002] Polarization synthetic aperture radar has many outstanding advantages, such as not being affected by time, and can image 24 hours a day. Polarization SAR images have unique advantages and broad application prospects, and have been successfully used in land use classification, change detection, surfac...

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): G06K9/62G06N3/04
CPCG06N3/045G06F18/241
Inventor 焦李成屈嵘高丽丽马文萍杨淑媛侯彪刘芳唐旭马晶晶张丹陈璞华古晶
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