A Bayesian framework for radar shape-from-shading

A G  Bors; E R  Hancock; R C  Wilson

A Bayesian framework for radar shape-from-shading

Computer Science

Research output: Contribution to journal › Article › peer-review

Abstract

This paper introduces a Bayesian approach to shape-from-shading (SFS) which is applied to terrain recovery in Synthetic Aperture Radar (SAR) images. The Bayesian model relates the recovery of 3-D shape information to the original 2-D radar intensity and to edges separating different topographic regions. First, we model the image amplitude distribution and the reflection function in SAR images. Using a maximum log-likelihood feature detector derived from the image statistics we identify the ridges and ravines in the terrain image. These topographic features are used to constrain the recovery of surface normals in the shape-from-shading process. Finally, the surface normals are smoothed using robust statistics operators.

Original language	English
Pages (from-to)	262-268
Number of pages	7
Journal	IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, VOL I
Publication status	Published - 2000

Keywords

IMAGES

Cite this

@article{c5d0711664ab4533951af0ef3bf93fce,

title = "A Bayesian framework for radar shape-from-shading",

abstract = "This paper introduces a Bayesian approach to shape-from-shading (SFS) which is applied to terrain recovery in Synthetic Aperture Radar (SAR) images. The Bayesian model relates the recovery of 3-D shape information to the original 2-D radar intensity and to edges separating different topographic regions. First, we model the image amplitude distribution and the reflection function in SAR images. Using a maximum log-likelihood feature detector derived from the image statistics we identify the ridges and ravines in the terrain image. These topographic features are used to constrain the recovery of surface normals in the shape-from-shading process. Finally, the surface normals are smoothed using robust statistics operators.",

keywords = "IMAGES",

author = "Bors, {A G} and Hancock, {E R} and Wilson, {R C}",

year = "2000",

language = "English",

pages = "262--268",

journal = "IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, VOL I",

issn = "1063-6919",

publisher = "IEEE Computer Society",

}

TY - JOUR

T1 - A Bayesian framework for radar shape-from-shading

AU - Bors, A G

AU - Hancock, E R

AU - Wilson, R C

PY - 2000

Y1 - 2000

N2 - This paper introduces a Bayesian approach to shape-from-shading (SFS) which is applied to terrain recovery in Synthetic Aperture Radar (SAR) images. The Bayesian model relates the recovery of 3-D shape information to the original 2-D radar intensity and to edges separating different topographic regions. First, we model the image amplitude distribution and the reflection function in SAR images. Using a maximum log-likelihood feature detector derived from the image statistics we identify the ridges and ravines in the terrain image. These topographic features are used to constrain the recovery of surface normals in the shape-from-shading process. Finally, the surface normals are smoothed using robust statistics operators.

AB - This paper introduces a Bayesian approach to shape-from-shading (SFS) which is applied to terrain recovery in Synthetic Aperture Radar (SAR) images. The Bayesian model relates the recovery of 3-D shape information to the original 2-D radar intensity and to edges separating different topographic regions. First, we model the image amplitude distribution and the reflection function in SAR images. Using a maximum log-likelihood feature detector derived from the image statistics we identify the ridges and ravines in the terrain image. These topographic features are used to constrain the recovery of surface normals in the shape-from-shading process. Finally, the surface normals are smoothed using robust statistics operators.

KW - IMAGES

M3 - Article

SP - 262

EP - 268

JO - IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, VOL I

JF - IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, VOL I

SN - 1063-6919

ER -