Temporally consistent snow cover estimation from noisy, irregularly sampled measurements

D. Rüfenacht, M. Brown, J. Beutel, S. Süsstrunk

Research output: Chapter in Book/Report/Conference proceedingChapter

4 Citations (Scopus)

Abstract

We propose a method for accurate and temporally consistent surface classification in the presence of noisy, irregularly sampled measurements, and apply it to the estimation of snow coverage over time. The input imagery is extremely challenging, with large variations in lighting and weather distorting the measurements. Initial snow cover estimations are obtained using a Gaussian Mixture Model of color. To achieve a temporally consistent snow cover estimation, we use a Markov Random Field that penalizes rapid fluctuations in the snow state, and show that the penalty term needs to be quite large, resulting in slow reactivity to changes. We thus propose a classifier to separate good from uninformative images, which allows to use a smaller penalty term. We show that the incorporation of domain knowledge to discard uninformative images leads to better reactivity to changes in snow coverage as well as more accurate snow cover estimations.
Original languageEnglish
Title of host publicationVISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications
Pages275-283
Number of pages9
Volume2
Publication statusPublished - 2014
Event9th International Conference on Computer Vision Theory and Applications, VISAPP, 2014 - Lisbon, Portugal
Duration: 5 Jan 20148 Jan 2014

Conference

Conference9th International Conference on Computer Vision Theory and Applications, VISAPP, 2014
CountryPortugal
CityLisbon
Period5/01/148/01/14

Fingerprint

snow cover
snow
imagery
weather
penalty
incorporation
method
lighting

Cite this

Rüfenacht, D., Brown, M., Beutel, J., & Süsstrunk, S. (2014). Temporally consistent snow cover estimation from noisy, irregularly sampled measurements. In VISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications (Vol. 2, pp. 275-283)

Temporally consistent snow cover estimation from noisy, irregularly sampled measurements. / Rüfenacht, D.; Brown, M.; Beutel, J.; Süsstrunk, S.

VISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications. Vol. 2 2014. p. 275-283.

Research output: Chapter in Book/Report/Conference proceedingChapter

Rüfenacht, D, Brown, M, Beutel, J & Süsstrunk, S 2014, Temporally consistent snow cover estimation from noisy, irregularly sampled measurements. in VISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications. vol. 2, pp. 275-283, 9th International Conference on Computer Vision Theory and Applications, VISAPP, 2014, Lisbon, Portugal, 5/01/14.
Rüfenacht D, Brown M, Beutel J, Süsstrunk S. Temporally consistent snow cover estimation from noisy, irregularly sampled measurements. In VISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications. Vol. 2. 2014. p. 275-283
Rüfenacht, D. ; Brown, M. ; Beutel, J. ; Süsstrunk, S. / Temporally consistent snow cover estimation from noisy, irregularly sampled measurements. VISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications. Vol. 2 2014. pp. 275-283
@inbook{41649abd7dd24f948b18ba427b3f3c6f,
title = "Temporally consistent snow cover estimation from noisy, irregularly sampled measurements",
abstract = "We propose a method for accurate and temporally consistent surface classification in the presence of noisy, irregularly sampled measurements, and apply it to the estimation of snow coverage over time. The input imagery is extremely challenging, with large variations in lighting and weather distorting the measurements. Initial snow cover estimations are obtained using a Gaussian Mixture Model of color. To achieve a temporally consistent snow cover estimation, we use a Markov Random Field that penalizes rapid fluctuations in the snow state, and show that the penalty term needs to be quite large, resulting in slow reactivity to changes. We thus propose a classifier to separate good from uninformative images, which allows to use a smaller penalty term. We show that the incorporation of domain knowledge to discard uninformative images leads to better reactivity to changes in snow coverage as well as more accurate snow cover estimations.",
author = "D. R{\"u}fenacht and M. Brown and J. Beutel and S. S{\"u}sstrunk",
year = "2014",
language = "English",
isbn = "9789897580048",
volume = "2",
pages = "275--283",
booktitle = "VISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications",

}

TY - CHAP

T1 - Temporally consistent snow cover estimation from noisy, irregularly sampled measurements

AU - Rüfenacht, D.

AU - Brown, M.

AU - Beutel, J.

AU - Süsstrunk, S.

PY - 2014

Y1 - 2014

N2 - We propose a method for accurate and temporally consistent surface classification in the presence of noisy, irregularly sampled measurements, and apply it to the estimation of snow coverage over time. The input imagery is extremely challenging, with large variations in lighting and weather distorting the measurements. Initial snow cover estimations are obtained using a Gaussian Mixture Model of color. To achieve a temporally consistent snow cover estimation, we use a Markov Random Field that penalizes rapid fluctuations in the snow state, and show that the penalty term needs to be quite large, resulting in slow reactivity to changes. We thus propose a classifier to separate good from uninformative images, which allows to use a smaller penalty term. We show that the incorporation of domain knowledge to discard uninformative images leads to better reactivity to changes in snow coverage as well as more accurate snow cover estimations.

AB - We propose a method for accurate and temporally consistent surface classification in the presence of noisy, irregularly sampled measurements, and apply it to the estimation of snow coverage over time. The input imagery is extremely challenging, with large variations in lighting and weather distorting the measurements. Initial snow cover estimations are obtained using a Gaussian Mixture Model of color. To achieve a temporally consistent snow cover estimation, we use a Markov Random Field that penalizes rapid fluctuations in the snow state, and show that the penalty term needs to be quite large, resulting in slow reactivity to changes. We thus propose a classifier to separate good from uninformative images, which allows to use a smaller penalty term. We show that the incorporation of domain knowledge to discard uninformative images leads to better reactivity to changes in snow coverage as well as more accurate snow cover estimations.

UR - http://www.scopus.com/inward/record.url?scp=84906919934&partnerID=8YFLogxK

M3 - Chapter

SN - 9789897580048

VL - 2

SP - 275

EP - 283

BT - VISAPP 2014 - Proceedings of the 9th International Conference on Computer Vision Theory and Applications

ER -