Polarized SAR image classification method based on semi-supervised depth distance metric network

A technology of depth distance and classification method, applied in the field of image processing, can solve the problems that the network performance cannot be guaranteed, the shallow features cannot fully reflect the intrinsic properties of the data, and the pre-training effect is not ideal, so as to improve the classification accuracy and improve The effect of lower classification accuracy

Active Publication Date: 2017-09-22
XIDIAN UNIV
View PDF3 Cites 29 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Traditional machine learning methods need to manually extract features and the obtained shallow features cannot fully reflect the intrinsic properties of the data
However, existing deep learning methods such as autoencoders and deep belief networks use unsupervised pre-training methods without label sample guidance. The ...

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 image classification method based on semi-supervised depth distance metric network
  • Polarized SAR image classification method based on semi-supervised depth distance metric network
  • Polarized SAR image classification method based on semi-supervised depth distance metric network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] Due to the development of remote sensing technology, it has been widely used in environmental monitoring, earth resource surveying, military systems and other fields, and the demand for polarimetric SAR image processing is also increasing. Deep learning has obvious advantages in machine learning methods. However, traditional deep learning networks are mostly unsupervised or supervised learning methods. When there are few labeled samples, unsupervised deep learning methods may underfit if they only rely on unsupervised pre-training, and a small number of labeled samples Fine-tuning cannot effectively improve network performance; for supervised deep learning methods, network training is insufficient and network performance is poor. Traditional deep learning methods do not take into account both the linear and nonlinear characteristics of samples, and the learned features cannot fully reflect the intrinsic properties of samples. In response to these current situations, the...

Embodiment 2

[0039] The polarized SAR image classification method based on the semi-supervised depth distance measurement network is the same as embodiment 1, and the Wishart nearest neighbor sample for each marked sample in step (3) includes the following steps:

[0040] 3a. The labeled sample matrix is Indicates the number of marked samples, and use the following formula to find the Wishart distance between each marked sample and the rest of the samples:

[0041] d(x i ,x j )=ln((x i ) -1 x j )+Tr((x j ) -1 x i )-q,

[0042] Among them, Tr () represents the trace of the matrix, for the radar whose transmission and reception are integrated, due to reciprocity, the constant q=3; for the radar whose transmission and reception are not integrated, the constant q=4;

[0043] 3b. Use the sort function in MATLAB to calculate the Wishart distance d(x i ,x j ) in ascending order of absolute value, taking the top K 1 labeled similar neighbor samples x j (j=1,2,...,K 1 ), K 2 unlab...

Embodiment 3

[0045] The polarized SAR image classification method based on the semi-supervised depth distance measurement network is the same as embodiment 1-2, and the process of constructing the loss function of the semi-supervised large boundary neighbor algorithm described in step (4) includes:

[0046] 4a. Find the loss function of the large boundary neighbor algorithm:

[0047] The distance square formula of the large boundary nearest neighbor algorithm is:

[0048]

[0049] Among them, L is a linear change matrix, x j is x i Similar labeled samples of . If there is a labeled sample x i x of non-homogeneous samples of l , satisfying the following formula:

[0050] ||L(x i -x l )|| 2 ≤||L(x i -x j )|| 2 +1,

[0051] Then, x l Known as an "imposter".

[0052] The large-boundary nearest neighbor algorithm can be expressed as two parts: the loss function ε between similar samples pull (L) and the loss function ε between non-similar samples push (L), ε pull (L) is 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
Login to view more

PUM

No PUM Login to view more

Abstract

The present invention discloses a polarized SAR image classification method based on the semi-supervised depth distance metric network, and the technical problems that the traditional depth learning only considers the non-linear relationship between the sample characteristics and the classification accuracy is not high when the number of marked samples is relatively small are solved. The method comprises the following steps: inputting to-be-classified polarized SAR image data; solving a neighboring sample of the marked sample; constructing the loss function of the semi-supervised large boundary neighbor algorithm; initializing parameters of the network; pre-training the network; carrying out fine tuning on the network; carrying out classification prediction on the unmarked samples; and outputting a classification result image and classification accuracy of the to-be-classified polarized SAR image. According to the method disclosed by the present invention, by constructing a depth distance metric network, a popular learning regular term is added to the large boundary neighbor algorithm, so that problems of the influence of insufficient marked samples on the classification accuracy and the waste of information of a large number of unmarked samples are overcome; and the characteristics learned in the method of the present invention fully depicts intrinsic attributes of the samples, and the method can be applied to the earth resources survey, military systems and other technical fields.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a polarization SAR image classification method, in particular to a polarization SAR image classification method based on a semi-supervised depth distance measurement network. It can be used in environmental monitoring, earth resource surveying and military systems, etc. Background technique [0002] Machine learning (Machine Learning, ML), as a subfield of computer science, starts from the research of artificial intelligence, computer learning theory and pattern recognition, and constructs an algorithm that can learn knowledge from data and predict similar data. Machine learning can learn its various attributes from the original data so as to have the ability to deal with various similar problems, which is how to make the computer automatically acquire new knowledge and new capabilities in experience learning. In the field of classification and recognition of polarimetr...

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): G06K9/00G06K9/62G06N99/00
CPCG06N20/00G06V20/13G06F18/22
Inventor 刘红英缑水平闵强焦李成熊涛冯婕侯彪王爽
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