Improved stereo matching in scattering media by incorporating a backscatter cue

Shahriar Negahdaripour, Amin Sarafraz

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

In scattering media, as in underwater or haze and fog in atmosphere, image contrast deteriorates significantly due to backscatter. This adversely affects the performance of many computer vision techniques developed for clear open-air conditions, including stereo matching, when applied to images acquired in these environments. Since the strength of the scattering depends on the distance to the scene points, the scattering field embodies range information that can be exploited for 3-D reconstruction. In this paper, we present an integrated solution for 3-D structure from stereovision that incorporates the visual cues from both disparity and scattering. The method applies to images of scenes illuminated by artificial sources and natural lighting, and performance improves with discrepancy between the backscatter fields in the two views. Neither source calibration nor knowledge of medium optical properties is required. Instead, backscatter fields at infinity, i.e., stereo images taken with no target in the field of view, are directly employed in the estimation process. Results from experiments with synthetic and real data demonstrate the key advantages of our method.

Original languageEnglish
Article number6901274
Pages (from-to)5743-5755
Number of pages13
JournalIEEE Transactions on Image Processing
Volume23
Issue number12
DOIs
StatePublished - Dec 1 2014

Fingerprint

Scattering
Fog
Computer vision
Optical properties
Lighting
Calibration
Experiments

Keywords

  • dense correspondence
  • scattering media
  • scene reconstruction
  • Stereo vision

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Improved stereo matching in scattering media by incorporating a backscatter cue. / Negahdaripour, Shahriar; Sarafraz, Amin.

In: IEEE Transactions on Image Processing, Vol. 23, No. 12, 6901274, 01.12.2014, p. 5743-5755.

Research output: Contribution to journalArticle

@article{a85768ebb980440cb1febed844c13c83,
title = "Improved stereo matching in scattering media by incorporating a backscatter cue",
abstract = "In scattering media, as in underwater or haze and fog in atmosphere, image contrast deteriorates significantly due to backscatter. This adversely affects the performance of many computer vision techniques developed for clear open-air conditions, including stereo matching, when applied to images acquired in these environments. Since the strength of the scattering depends on the distance to the scene points, the scattering field embodies range information that can be exploited for 3-D reconstruction. In this paper, we present an integrated solution for 3-D structure from stereovision that incorporates the visual cues from both disparity and scattering. The method applies to images of scenes illuminated by artificial sources and natural lighting, and performance improves with discrepancy between the backscatter fields in the two views. Neither source calibration nor knowledge of medium optical properties is required. Instead, backscatter fields at infinity, i.e., stereo images taken with no target in the field of view, are directly employed in the estimation process. Results from experiments with synthetic and real data demonstrate the key advantages of our method.",
keywords = "dense correspondence, scattering media, scene reconstruction, Stereo vision",
author = "Shahriar Negahdaripour and Amin Sarafraz",
year = "2014",
month = "12",
day = "1",
doi = "10.1109/TIP.2014.2358882",
language = "English",
volume = "23",
pages = "5743--5755",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

TY - JOUR

T1 - Improved stereo matching in scattering media by incorporating a backscatter cue

AU - Negahdaripour, Shahriar

AU - Sarafraz, Amin

PY - 2014/12/1

Y1 - 2014/12/1

N2 - In scattering media, as in underwater or haze and fog in atmosphere, image contrast deteriorates significantly due to backscatter. This adversely affects the performance of many computer vision techniques developed for clear open-air conditions, including stereo matching, when applied to images acquired in these environments. Since the strength of the scattering depends on the distance to the scene points, the scattering field embodies range information that can be exploited for 3-D reconstruction. In this paper, we present an integrated solution for 3-D structure from stereovision that incorporates the visual cues from both disparity and scattering. The method applies to images of scenes illuminated by artificial sources and natural lighting, and performance improves with discrepancy between the backscatter fields in the two views. Neither source calibration nor knowledge of medium optical properties is required. Instead, backscatter fields at infinity, i.e., stereo images taken with no target in the field of view, are directly employed in the estimation process. Results from experiments with synthetic and real data demonstrate the key advantages of our method.

AB - In scattering media, as in underwater or haze and fog in atmosphere, image contrast deteriorates significantly due to backscatter. This adversely affects the performance of many computer vision techniques developed for clear open-air conditions, including stereo matching, when applied to images acquired in these environments. Since the strength of the scattering depends on the distance to the scene points, the scattering field embodies range information that can be exploited for 3-D reconstruction. In this paper, we present an integrated solution for 3-D structure from stereovision that incorporates the visual cues from both disparity and scattering. The method applies to images of scenes illuminated by artificial sources and natural lighting, and performance improves with discrepancy between the backscatter fields in the two views. Neither source calibration nor knowledge of medium optical properties is required. Instead, backscatter fields at infinity, i.e., stereo images taken with no target in the field of view, are directly employed in the estimation process. Results from experiments with synthetic and real data demonstrate the key advantages of our method.

KW - dense correspondence

KW - scattering media

KW - scene reconstruction

KW - Stereo vision

UR - http://www.scopus.com/inward/record.url?scp=84913554829&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84913554829&partnerID=8YFLogxK

U2 - 10.1109/TIP.2014.2358882

DO - 10.1109/TIP.2014.2358882

M3 - Article

AN - SCOPUS:84913554829

VL - 23

SP - 5743

EP - 5755

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 12

M1 - 6901274

ER -