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

Method for segmenting heterogeneous super-pixel SAR (Synthetic Aperture Radar) image based on Gamma distribution

An image segmentation and super-pixel technology, applied in the field of image processing, can solve the problems of reducing the resolution and image quality of SAR images, the consistency of mis-segmented areas, and unsatisfactory, etc., to achieve easy maximum expectation method processing, suppress coherent speckle noise, The effect of close correspondence

Active Publication Date: 2015-11-25
XIDIAN UNIV
View PDF2 Cites 23 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the inherent coherent speckle noise in SAR images, the edge information of segmented objects is destroyed, and the resolution and image quality of SAR images are reduced at the same time.
The existence of speckle noise seriously affects the segmentation performance of IRGS algorithm and RCBLP algorithm on SAR images, and it will cause mis-segmentation and unsatisfactory regional consistency for SAR images containing complex textures.

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
  • Method for segmenting heterogeneous super-pixel SAR (Synthetic Aperture Radar) image based on Gamma distribution
  • Method for segmenting heterogeneous super-pixel SAR (Synthetic Aperture Radar) image based on Gamma distribution
  • Method for segmenting heterogeneous super-pixel SAR (Synthetic Aperture Radar) image based on Gamma distribution

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] The present invention is a heterogeneous superpixel SAR image segmentation method based on Gamma distribution, see figure 1 , the specific implementation includes the following steps:

[0037] 1) Perform superpixel pre-segmentation on the image M*N to obtain S superpixel blocks. In this example will figure 2 (a) A SAR image with a size of 254*255, one of which is rivers and the other is land, is divided into S original superpixel blocks by using the turbopixel superpixel method.

[0038] 2) After superpixel pre-segmentation, use Gamma distribution to estimate the heterogeneity of each superpixel block, and take the heterogeneity parameter threshold m of Gamma distribution as the boundary to distinguish homogeneous superpixel blocks from heterogeneous blocks open, and reclassify the heterogeneous superpixel block with the Kmeans method; when the heterogeneity parameter is greater than the threshold m, the superpixel is a homogeneous superpixel, that is, the superpixel...

Embodiment 2

[0050] The heterogeneous superpixel SAR image segmentation method based on Gamma distribution is the same as that in Embodiment 1, and this example is for image 3 (a) Segmentation, wherein the re-segmentation after the superpixel pre-segmentation described in step 2 is specifically:

[0051] Use the Gamma distribution to estimate the heterogeneity of superpixels, and find out the superpixel blocks that are segmented incorrectly due to the weak boundary of the superpixels. If the superpixel block is too small, that is, it contains too few pixels, it does not need to be estimated. The Gamma distribution is defined as follows:

[0052] p ( R ) = 1 Γ ( υ ) [ υ E ( R ...

Embodiment 3

[0058] The heterogeneous superpixel SAR image segmentation method based on Gamma distribution is the same as that in Embodiment 1-2. In this example, 4(a) is segmented, and the heterogeneity parameter threshold m estimated by Gamma is set to 0.3.

[0059] The smaller m is, the more heterogeneous blocks can be obtained, and more details can be obtained. However, because the SAR image is affected by noise, if m is too small, homogeneous blocks of superpixels will be misclassified; if m is greater than 0.3, find After a large number of experimental analysis, research, analysis and comparison of superpixel heterogeneous blocks that cannot be classified incorrectly, the range of the heterogeneity parameter threshold m value given by the present invention is most suitable between 0.15 and 0.3.

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 method for segmenting a heterogeneous super-pixel SAR (Synthetic Aperture Radar) image based on Gamma distribution, used for mainly solving the problems of incomplete detailed information and disordered boundaries easily due to super-pixel pre-segmentation of the SAR image. The method comprises the following steps: performing super-pixel pre-segmentation of the image; estimating the heterogeneity of each super-pixel by using Gamma distribution, and finding out a wrongly segmented super-pixel block due to the weak boundary of the super-pixel; dividing wrongly segmented super-pixels into two kinds by using Kmeans; extracting features of each point in the original image, and extracting image texture features by adopting a gray co-occurrence matrix so as to obtain four-dimensional gray co-occurrence matrix features; extracting wavelet features of the original image so as to obtain nine-dimensional gabor features; distributing equal weights to the image texture features and scattering features, and fusing feature combinations; averaging pixel features in the super-pixel block so as to obtain features of each super-pixel; and classifying by using a Gaussian mixture model so as to obtain a final segmentation result. According to the invention, heterogeneous super-pixels can be segmented effectively; complete detailed information is reserved; the consistency of homogeneous regions is good; and noise is effectively inhibited.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a method for SAR image segmentation, in particular to a Gamma distribution-based heterogeneous superpixel SAR image segmentation method, which can be applied to SAR image segmentation. Background technique [0002] Synthetic Aperture Radar (SAR) is a high-resolution radar system that can be used in many fields such as military affairs, agriculture, navigation, and geographic surveillance. The SAR image is a representation of the scattering characteristics of radar waves, and it is a reflection of ground objects. The speckle noise in the image appears on a uniform surface, and the pixels appear as dark spots or bright spots. In fact, they are It can be seen that the gray pixel value of the image changes sharply, which reduces the spatial resolution of the image, blurs the edge information of the image, and reduces the accuracy of interpreting the image. [0003] Image se...

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): G06T7/00
CPCG06T2207/10044
Inventor 侯彪焦李成许声红马晶晶熊涛马文萍刘红英
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