A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence Set
The success of many computer vision and pattern recognition applications depends on matching local features on two or more images. Because the initial correspondence set—i.e., the set of the initial feature pairs—is often contaminated by mismatches, removing mismatches is a necessary task prior to i...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-07-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/14/14/3256 |
_version_ | 1797433107777847296 |
---|---|
author | Kuo-Liang Chung Ya-Chi Tseng Hsuan-Ying Chen |
author_facet | Kuo-Liang Chung Ya-Chi Tseng Hsuan-Ying Chen |
author_sort | Kuo-Liang Chung |
collection | DOAJ |
description | The success of many computer vision and pattern recognition applications depends on matching local features on two or more images. Because the initial correspondence set—i.e., the set of the initial feature pairs—is often contaminated by mismatches, removing mismatches is a necessary task prior to image matching. In this paper, we first propose a fast geometry histogram-based (GH-based) mismatch removal strategy to construct a reduced correspondence set <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>r</mi><mi>e</mi><mi>d</mi><mi>u</mi><mi>c</mi><mi>e</mi><mi>d</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula> from the initial correspondence set <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>. Next, we propose an effective cooperative random sample consensus (COOSAC) method for remote sensing image matching. COOSAC consists of a RANSAC, called <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></mrow></semantics></math></inline-formula> working on <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>, and a tiny RANSAC, called <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></mrow></semantics></math></inline-formula> working on a randomly selected subset of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>r</mi><mi>e</mi><mi>d</mi><mi>u</mi><mi>c</mi><mi>e</mi><mi>d</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula>. In <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></mrow></semantics></math></inline-formula>, an iterative area constraint-based sampling strategy is proposed to estimate the model solution of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula> until the specified confidence level is reached, and then <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></mrow></semantics></math></inline-formula> utilizes the estimated model solution of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula> to calculate the inlier rate of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>. COOSAC repeats the above cooperation between <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></mrow></semantics></math></inline-formula> until the specified confidence level is reached, reporting the resultant model solution of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>. For convenience, our image matching method is called the GH-COOSAC method. Based on several testing datasets, thorough experimental results demonstrate that the proposed GH-COOSAC method achieves lower computational cost and higher matching accuracy benefits when compared with the state-of-the-art image matching methods. |
first_indexed | 2024-03-09T10:12:25Z |
format | Article |
id | doaj.art-5113b5f343c84f36b29d7a3fa8bb89aa |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-09T10:12:25Z |
publishDate | 2022-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-5113b5f343c84f36b29d7a3fa8bb89aa2023-12-01T22:38:28ZengMDPI AGRemote Sensing2072-42922022-07-011414325610.3390/rs14143256A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence SetKuo-Liang Chung0Ya-Chi Tseng1Hsuan-Ying Chen2Department of Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taipei 10672, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taipei 10672, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taipei 10672, TaiwanThe success of many computer vision and pattern recognition applications depends on matching local features on two or more images. Because the initial correspondence set—i.e., the set of the initial feature pairs—is often contaminated by mismatches, removing mismatches is a necessary task prior to image matching. In this paper, we first propose a fast geometry histogram-based (GH-based) mismatch removal strategy to construct a reduced correspondence set <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>r</mi><mi>e</mi><mi>d</mi><mi>u</mi><mi>c</mi><mi>e</mi><mi>d</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula> from the initial correspondence set <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>. Next, we propose an effective cooperative random sample consensus (COOSAC) method for remote sensing image matching. COOSAC consists of a RANSAC, called <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></mrow></semantics></math></inline-formula> working on <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>, and a tiny RANSAC, called <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></mrow></semantics></math></inline-formula> working on a randomly selected subset of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>r</mi><mi>e</mi><mi>d</mi><mi>u</mi><mi>c</mi><mi>e</mi><mi>d</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula>. In <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></mrow></semantics></math></inline-formula>, an iterative area constraint-based sampling strategy is proposed to estimate the model solution of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula> until the specified confidence level is reached, and then <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></mrow></semantics></math></inline-formula> utilizes the estimated model solution of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></semantics></math></inline-formula> to calculate the inlier rate of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>. COOSAC repeats the above cooperation between <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>t</mi><mi>i</mi><mi>n</mi><mi>y</mi><mo>,</mo><mi>G</mi><mi>H</mi></mrow></msup></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><mi>A</mi><mi>N</mi><mi>S</mi><mi>A</mi><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></mrow></semantics></math></inline-formula> until the specified confidence level is reached, reporting the resultant model solution of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>C</mi><mrow><mi>i</mi><mi>n</mi><mi>i</mi></mrow></msup></semantics></math></inline-formula>. For convenience, our image matching method is called the GH-COOSAC method. Based on several testing datasets, thorough experimental results demonstrate that the proposed GH-COOSAC method achieves lower computational cost and higher matching accuracy benefits when compared with the state-of-the-art image matching methods.https://www.mdpi.com/2072-4292/14/14/3256confidence levelcomputational costhypothesize-and-verifyinlier ratematching accuracymodel solution |
spellingShingle | Kuo-Liang Chung Ya-Chi Tseng Hsuan-Ying Chen A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence Set Remote Sensing confidence level computational cost hypothesize-and-verify inlier rate matching accuracy model solution |
title | A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence Set |
title_full | A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence Set |
title_fullStr | A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence Set |
title_full_unstemmed | A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence Set |
title_short | A Novel and Effective Cooperative RANSAC Image Matching Method Using Geometry Histogram-Based Constructed Reduced Correspondence Set |
title_sort | novel and effective cooperative ransac image matching method using geometry histogram based constructed reduced correspondence set |
topic | confidence level computational cost hypothesize-and-verify inlier rate matching accuracy model solution |
url | https://www.mdpi.com/2072-4292/14/14/3256 |
work_keys_str_mv | AT kuoliangchung anovelandeffectivecooperativeransacimagematchingmethodusinggeometryhistogrambasedconstructedreducedcorrespondenceset AT yachitseng anovelandeffectivecooperativeransacimagematchingmethodusinggeometryhistogrambasedconstructedreducedcorrespondenceset AT hsuanyingchen anovelandeffectivecooperativeransacimagematchingmethodusinggeometryhistogrambasedconstructedreducedcorrespondenceset AT kuoliangchung novelandeffectivecooperativeransacimagematchingmethodusinggeometryhistogrambasedconstructedreducedcorrespondenceset AT yachitseng novelandeffectivecooperativeransacimagematchingmethodusinggeometryhistogrambasedconstructedreducedcorrespondenceset AT hsuanyingchen novelandeffectivecooperativeransacimagematchingmethodusinggeometryhistogrambasedconstructedreducedcorrespondenceset |