Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier Detection
In this paper, a robust technique based on a genetic algorithm is proposed for estimating two-view epipolar-geometry of uncalibrated perspective stereo images from putative correspondences containing a high percentage of outliers. The advantages of this technique are three-fold: (i) replacing random...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-07-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/10/3/87 |
| _version_ | 1819181500845785088 |
|---|---|
| author | Mozhdeh Shahbazi Gunho Sohn Jérôme Théau |
| author_facet | Mozhdeh Shahbazi Gunho Sohn Jérôme Théau |
| author_sort | Mozhdeh Shahbazi |
| collection | DOAJ |
| description | In this paper, a robust technique based on a genetic algorithm is proposed for estimating two-view epipolar-geometry of uncalibrated perspective stereo images from putative correspondences containing a high percentage of outliers. The advantages of this technique are three-fold: (i) replacing random search with evolutionary search applying new strategies of encoding and guided sampling; (ii) robust and fast estimation of the epipolar geometry via detecting a more-than-enough set of inliers without making any assumptions about the probability distribution of the residuals; (iii) determining the inlier-outlier threshold based on the uncertainty of the estimated model. The proposed method was evaluated both on synthetic data and real images. The results were compared with the most popular techniques from the state-of-the-art, including RANSAC (random sample consensus), MSAC, MLESAC, Cov-RANSAC, LO-RANSAC, StaRSAC, Multi-GS RANSAC and least median of squares (LMedS). Experimental results showed that the proposed approach performed better than other methods regarding the accuracy of inlier detection and epipolar-geometry estimation, as well as the computational efficiency for datasets majorly contaminated by outliers and noise. |
| first_indexed | 2024-12-22T22:31:14Z |
| format | Article |
| id | doaj.art-b5cf6784f3404c24b10f6b909e5b6b60 |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-12-22T22:31:14Z |
| publishDate | 2017-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-b5cf6784f3404c24b10f6b909e5b6b602022-12-21T18:10:27ZengMDPI AGAlgorithms1999-48932017-07-011038710.3390/a10030087a10030087Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier DetectionMozhdeh Shahbazi0Gunho Sohn1Jérôme Théau2Department of Geomatics Engineering, University of Calgary, 2500 University Drive NW, Calgary, AB T2N 1N4, CanadaDepartment of Earth and Space Science and Engineering, York University, 4700 Keele Street, Toronto, ON M3J 1P3, CanadaDepartment of Applied Geomatics, Université de Sherbrooke, 2500 Boulevard de l’Université, Sherbrooke, QC J1K 2R1, CanadaIn this paper, a robust technique based on a genetic algorithm is proposed for estimating two-view epipolar-geometry of uncalibrated perspective stereo images from putative correspondences containing a high percentage of outliers. The advantages of this technique are three-fold: (i) replacing random search with evolutionary search applying new strategies of encoding and guided sampling; (ii) robust and fast estimation of the epipolar geometry via detecting a more-than-enough set of inliers without making any assumptions about the probability distribution of the residuals; (iii) determining the inlier-outlier threshold based on the uncertainty of the estimated model. The proposed method was evaluated both on synthetic data and real images. The results were compared with the most popular techniques from the state-of-the-art, including RANSAC (random sample consensus), MSAC, MLESAC, Cov-RANSAC, LO-RANSAC, StaRSAC, Multi-GS RANSAC and least median of squares (LMedS). Experimental results showed that the proposed approach performed better than other methods regarding the accuracy of inlier detection and epipolar-geometry estimation, as well as the computational efficiency for datasets majorly contaminated by outliers and noise.https://www.mdpi.com/1999-4893/10/3/87sparse matchingoutlier detectiongenetic algorithmepipolar geometryevolutionary searchguided samplingadaptive thresholding |
| spellingShingle | Mozhdeh Shahbazi Gunho Sohn Jérôme Théau Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier Detection Algorithms sparse matching outlier detection genetic algorithm epipolar geometry evolutionary search guided sampling adaptive thresholding |
| title | Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier Detection |
| title_full | Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier Detection |
| title_fullStr | Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier Detection |
| title_full_unstemmed | Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier Detection |
| title_short | Evolutionary Optimization for Robust Epipolar-Geometry Estimation and Outlier Detection |
| title_sort | evolutionary optimization for robust epipolar geometry estimation and outlier detection |
| topic | sparse matching outlier detection genetic algorithm epipolar geometry evolutionary search guided sampling adaptive thresholding |
| url | https://www.mdpi.com/1999-4893/10/3/87 |
| work_keys_str_mv | AT mozhdehshahbazi evolutionaryoptimizationforrobustepipolargeometryestimationandoutlierdetection AT gunhosohn evolutionaryoptimizationforrobustepipolargeometryestimationandoutlierdetection AT jerometheau evolutionaryoptimizationforrobustepipolargeometryestimationandoutlierdetection |