3-D mapping of sea floor scenes by stereo imaging

A. Khamene, H. Madjidi, Shahriar Negahdaripour

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

We investigate the recovery of 3D motion and structure from the stereo images of a stationary environment. A Kalman filter-based framework is proposed for the reconstruction of 3D structure from multiple visual cues, through the integration of image motion and stereo disparity with the shading flow that is induced by the rotational motion of the source. This allows the exploitation of available visual cues in the common scenario involving the coupled motion of artificial source(s) and stereo cameras that are installed on mobile submersible vehicles. Utilizing shading flow with the image motion leads to devising a more robust 3D motion estimation algorithm, in addition to the critical role in depth recovery/refinement by constraining the local surface gradients. Collectively, use of multiple cues enhances robustness with respect to perturbation in any of the cues. Results of experiments with real imagery are presented to evaluate the performance of the proposed algorithm.

Original languageEnglish (US)
Title of host publicationOceans Conference Record (IEEE)
Pages2577-2581
Number of pages5
Volume4
StatePublished - 2001
EventOceans 2001 MTS/IEEE - An Ocean Odyssey - Honolulu, HI, United States
Duration: Nov 5 2001Nov 8 2001

Other

OtherOceans 2001 MTS/IEEE - An Ocean Odyssey
CountryUnited States
CityHonolulu, HI
Period11/5/0111/8/01

Fingerprint

visual cue
shading
seafloor
stereo image
submersible
Kalman filter
imagery
perturbation
experiment
vehicle

ASJC Scopus subject areas

  • Oceanography

Cite this

Khamene, A., Madjidi, H., & Negahdaripour, S. (2001). 3-D mapping of sea floor scenes by stereo imaging. In Oceans Conference Record (IEEE) (Vol. 4, pp. 2577-2581)

3-D mapping of sea floor scenes by stereo imaging. / Khamene, A.; Madjidi, H.; Negahdaripour, Shahriar.

Oceans Conference Record (IEEE). Vol. 4 2001. p. 2577-2581.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Khamene, A, Madjidi, H & Negahdaripour, S 2001, 3-D mapping of sea floor scenes by stereo imaging. in Oceans Conference Record (IEEE). vol. 4, pp. 2577-2581, Oceans 2001 MTS/IEEE - An Ocean Odyssey, Honolulu, HI, United States, 11/5/01.
Khamene A, Madjidi H, Negahdaripour S. 3-D mapping of sea floor scenes by stereo imaging. In Oceans Conference Record (IEEE). Vol. 4. 2001. p. 2577-2581
Khamene, A. ; Madjidi, H. ; Negahdaripour, Shahriar. / 3-D mapping of sea floor scenes by stereo imaging. Oceans Conference Record (IEEE). Vol. 4 2001. pp. 2577-2581
@inproceedings{43a828ba959c4176896d7ef1b923bcbc,
title = "3-D mapping of sea floor scenes by stereo imaging",
abstract = "We investigate the recovery of 3D motion and structure from the stereo images of a stationary environment. A Kalman filter-based framework is proposed for the reconstruction of 3D structure from multiple visual cues, through the integration of image motion and stereo disparity with the shading flow that is induced by the rotational motion of the source. This allows the exploitation of available visual cues in the common scenario involving the coupled motion of artificial source(s) and stereo cameras that are installed on mobile submersible vehicles. Utilizing shading flow with the image motion leads to devising a more robust 3D motion estimation algorithm, in addition to the critical role in depth recovery/refinement by constraining the local surface gradients. Collectively, use of multiple cues enhances robustness with respect to perturbation in any of the cues. Results of experiments with real imagery are presented to evaluate the performance of the proposed algorithm.",
author = "A. Khamene and H. Madjidi and Shahriar Negahdaripour",
year = "2001",
language = "English (US)",
volume = "4",
pages = "2577--2581",
booktitle = "Oceans Conference Record (IEEE)",

}

TY - GEN

T1 - 3-D mapping of sea floor scenes by stereo imaging

AU - Khamene, A.

AU - Madjidi, H.

AU - Negahdaripour, Shahriar

PY - 2001

Y1 - 2001

N2 - We investigate the recovery of 3D motion and structure from the stereo images of a stationary environment. A Kalman filter-based framework is proposed for the reconstruction of 3D structure from multiple visual cues, through the integration of image motion and stereo disparity with the shading flow that is induced by the rotational motion of the source. This allows the exploitation of available visual cues in the common scenario involving the coupled motion of artificial source(s) and stereo cameras that are installed on mobile submersible vehicles. Utilizing shading flow with the image motion leads to devising a more robust 3D motion estimation algorithm, in addition to the critical role in depth recovery/refinement by constraining the local surface gradients. Collectively, use of multiple cues enhances robustness with respect to perturbation in any of the cues. Results of experiments with real imagery are presented to evaluate the performance of the proposed algorithm.

AB - We investigate the recovery of 3D motion and structure from the stereo images of a stationary environment. A Kalman filter-based framework is proposed for the reconstruction of 3D structure from multiple visual cues, through the integration of image motion and stereo disparity with the shading flow that is induced by the rotational motion of the source. This allows the exploitation of available visual cues in the common scenario involving the coupled motion of artificial source(s) and stereo cameras that are installed on mobile submersible vehicles. Utilizing shading flow with the image motion leads to devising a more robust 3D motion estimation algorithm, in addition to the critical role in depth recovery/refinement by constraining the local surface gradients. Collectively, use of multiple cues enhances robustness with respect to perturbation in any of the cues. Results of experiments with real imagery are presented to evaluate the performance of the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=0035675806&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035675806&partnerID=8YFLogxK

M3 - Conference contribution

VL - 4

SP - 2577

EP - 2581

BT - Oceans Conference Record (IEEE)

ER -