More Web Proxy on the site http://driver.im/

You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.

Search for Articles:

Title / Keyword

Author / Affiliation / Email

Journal

Article Type

Advanced Search

Section

Special Issue

Volume

Issue

Number

Page

Logical OperatorOperator

Search Text

Search Type

Search Results (1)

Search Parameters:
Keywords = RHLBP

Order results

Result details

Results per page

Show export options Show export options

Select all

Export citation of selected articles as:

5272 KiB

Open AccessArticle

Segmentation-Based PolSAR Image Classification Using Visual Features: RHLBP and Color Features

by Jian Cheng, Yaqi Ji and Haijun Liu

Remote Sens. 2015, 7(5), 6079-6106; https://doi.org/10.3390/rs70506079 - 15 May 2015

Cited by 25 | Viewed by 5997

A segmentation-based fully-polarimetric synthetic aperture radar (PolSAR) image classification method that incorporates texture features and color features is designed and implemented. This method is based on the framework that conjunctively uses statistical region merging (SRM) for segmentation and support vector machine (SVM) for [...] Read more.

A segmentation-based fully-polarimetric synthetic aperture radar (PolSAR) image classification method that incorporates texture features and color features is designed and implemented. This method is based on the framework that conjunctively uses statistical region merging (SRM) for segmentation and support vector machine (SVM) for classification. In the segmentation step, we propose an improved local binary pattern (LBP) operator named the regional homogeneity local binary pattern (RHLBP) to guarantee the regional homogeneity in PolSAR images. In the classification step, the color features extracted from false color images are applied to improve the classification accuracy. The RHLBP operator and color features can provide discriminative information to separate those pixels and regions with similar polarimetric features, which are from different classes. Extensive experimental comparison results with conventional methods on L-band PolSAR data demonstrate the effectiveness of our proposed method for PolSAR image classification. Full article

► Show Figures

<p>The procedure of the proposed PolSAR image classification algorithm. (<b>a</b>) a false color image, (<b>b</b>) a segmented image; (<b>c</b>) a classification result. (Note: the difference between (a) and (b) is that (a) is a pixel-based image with speckle noise, while (b) is a region-based image without speckle noise.)</p> Full article ">
<p>Two pixels belonging to the same class and their binarization results. (a,c) The results by rotation invariant uniform local binary pattern (RIU-LBP); (b,d) the results by regional homogeneity local binary pattern (RHLBP) with <span class="html-italic">T</span> = 20.</p> Full article ">
<p>Two pixels belonging to different classes and their binarization results. (a,c) The results by RIU-LBP; (b,d) the results by RHLBP with <span class="html-italic">T</span> = 20.</p> Full article ">
<p>Tailored regions: (<b>a</b>) heterogeneous regions; (<b>b</b>) homogenous regions.</p> Full article ">
<p>The <math display="inline"> <mrow> <mi>L</mi> <mi>B</mi> <msubsup> <mi>P</mi> <mrow> <mn>8</mn> <mo>,</mo> <mn>1</mn></mrow> <mrow> <mi>r</mi> <mi>i</mi> <mi>u</mi> <mn>2</mn></mrow></msubsup></mrow></math> and <math display="inline"> <mrow> <mi>L</mi> <mi>B</mi> <msubsup> <mi>P</mi> <mrow> <mn>8</mn> <mo>,</mo> <mn>1</mn></mrow> <mrow> <mi>r</mi> <mi>i</mi> <mi>u</mi> <mn>2</mn> <mo>,</mo> <mi>T</mi></mrow></msubsup></mrow></math> value distributions of the heterogeneous regions in <a href="#f4-remotesensing-07-06079" class="html-fig">Figure 4a</a>: (<b>a</b>) RIU-LBP; (<b>b</b>) RHLBP (T = 10); (<b>c</b>) RHLBP (T = 20); (<b>d</b>) RHLBP (T = 30).</p> Full article ">
<p>The <math display="inline"> <mrow> <mi>L</mi> <mi>B</mi> <msubsup> <mi>P</mi> <mrow> <mn>8</mn> <mo>,</mo> <mn>1</mn></mrow> <mrow> <mi>r</mi> <mi>i</mi> <mi>u</mi> <mn>2</mn></mrow></msubsup></mrow></math> and <math display="inline"> <mrow> <mi>L</mi> <mi>B</mi> <msubsup> <mi>P</mi> <mrow> <mn>8</mn> <mo>,</mo> <mn>1</mn></mrow> <mrow> <mi>r</mi> <mi>i</mi> <mi>u</mi> <mn>2</mn> <mo>,</mo> <mi>T</mi></mrow></msubsup></mrow></math> value distributions of the homogenous regions in <a href="#f4-remotesensing-07-06079" class="html-fig">Figure 4b</a>: (<b>a</b>) RIU-LBP; (<b>b</b>) RHLBP (T = 10); (<b>c</b>) RHLBP (T = 20); (<b>d</b>) RHLBP (T = 30).</p> Full article ">
<p>The Bhattacharyya distances of the two pairs of regions in <a href="#f4-remotesensing-07-06079" class="html-fig">Figure 4</a>: (<b>a</b>) heterogeneous regions; (<b>b</b>) homogenous regions.</p> Full article ">
<p>(Left) The false color image of the Flevoland image and (right) the ground truth image, where the black regions are without the accurate class information.</p> Full article ">
<p>The segmentation results of the Flevoland image using the statistical region merging (SRM) algorithm: (<b>a</b>) Q = 100; (<b>b</b>) Q = 500.</p> Full article ">

Show export options Show export options

Select all

Export citation of selected articles as:

Displaying article 1-50 on page 1 of 1.

Back to TopTop