Change detection in bi-temporal data by canonical information analysis

Allan Aasbjerg Nielsen; Jacob Schack Vestergaard

doi:10.1109/Multi-Temp.2015.7245779

Change detection in bi-temporal data by canonical information analysis

Allan Aasbjerg Nielsen
, Jacob Schack Vestergaard

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Abstract

Canonical correlation analysis (CCA) is an established multivariate statistical method for finding similarities between linear combinations of (normally two) sets of multivariate observations. In this contribution we replace (linear) correlation as the measure of association between the linear combinations with the information theoretical measure mutual information (MI). We term this type of analysis canonical information analysis (CIA). MI allows for the actual joint distribution of the variables involved and not just second order statistics. Where CCA is ideal for Gaussian data, CIA facilitates analysis of variables with different genesis and therefore different statistical distributions. As a proof of concept we give a toy example. We also give an example with DLR 3K camera data from two time points covering a motor way.

Original language	English
Title of host publication	Proceedings of the 8th International Workshop on the Analysis of Multitemporal Remote Sensing Images (Multi-Temp 2015)
Number of pages	4
Publisher	IEEE
Publication date	2015
ISBN (Print)	978-1-4673-7119-3
DOIs	https://doi.org/10.1109/Multi-Temp.2015.7245779
Publication status	Published - 2015
Event	8th International Workshop on the Analysis of Multitemporal Remote Sensing Images - Annecy, France Duration: 22 Jul 2015 → 24 Jul 2015 Conference number: 8

Workshop

Workshop	8th International Workshop on the Analysis of Multitemporal Remote Sensing Images
Number	8
Country/Territory	France
City	Annecy
Period	22/07/2015 → 24/07/2015

Access to Document

10.1109/Multi-Temp.2015.7245779

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{aea144479ab14fc1abfdeb03f693f0e9,

title = "Change detection in bi-temporal data by canonical information analysis",

abstract = "Canonical correlation analysis (CCA) is an established multivariate statistical method for finding similarities between linear combinations of (normally two) sets of multivariate observations. In this contribution we replace (linear) correlation as the measure of association between the linear combinations with the information theoretical measure mutual information (MI). We term this type of analysis canonical information analysis (CIA). MI allows for the actual joint distribution of the variables involved and not just second order statistics. Where CCA is ideal for Gaussian data, CIA facilitates analysis of variables with different genesis and therefore different statistical distributions. As a proof of concept we give a toy example. We also give an example with DLR 3K camera data from two time points covering a motor way.",

author = "Nielsen, \{Allan Aasbjerg\} and Vestergaard, \{Jacob Schack\}",

year = "2015",

doi = "10.1109/Multi-Temp.2015.7245779",

language = "English",

isbn = "978-1-4673-7119-3",

booktitle = "Proceedings of the 8th International Workshop on the Analysis of Multitemporal Remote Sensing Images (Multi-Temp 2015)",

publisher = "IEEE",

address = "United States",

note = "8th International Workshop on the Analysis of Multitemporal Remote Sensing Images, Multi-Temp 2015 ; Conference date: 22-07-2015 Through 24-07-2015",

}

Nielsen, AA & Vestergaard, JS 2015, Change detection in bi-temporal data by canonical information analysis. in Proceedings of the 8th International Workshop on the Analysis of Multitemporal Remote Sensing Images (Multi-Temp 2015). IEEE, 8th International Workshop on the Analysis of Multitemporal Remote Sensing Images, Annecy, France, 22/07/2015. https://doi.org/10.1109/Multi-Temp.2015.7245779

Change detection in bi-temporal data by canonical information analysis. / Nielsen, Allan Aasbjerg; Vestergaard, Jacob Schack.
Proceedings of the 8th International Workshop on the Analysis of Multitemporal Remote Sensing Images (Multi-Temp 2015). IEEE, 2015.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Change detection in bi-temporal data by canonical information analysis

AU - Nielsen, Allan Aasbjerg

AU - Vestergaard, Jacob Schack

N1 - Conference code: 8

PY - 2015

Y1 - 2015

N2 - Canonical correlation analysis (CCA) is an established multivariate statistical method for finding similarities between linear combinations of (normally two) sets of multivariate observations. In this contribution we replace (linear) correlation as the measure of association between the linear combinations with the information theoretical measure mutual information (MI). We term this type of analysis canonical information analysis (CIA). MI allows for the actual joint distribution of the variables involved and not just second order statistics. Where CCA is ideal for Gaussian data, CIA facilitates analysis of variables with different genesis and therefore different statistical distributions. As a proof of concept we give a toy example. We also give an example with DLR 3K camera data from two time points covering a motor way.

AB - Canonical correlation analysis (CCA) is an established multivariate statistical method for finding similarities between linear combinations of (normally two) sets of multivariate observations. In this contribution we replace (linear) correlation as the measure of association between the linear combinations with the information theoretical measure mutual information (MI). We term this type of analysis canonical information analysis (CIA). MI allows for the actual joint distribution of the variables involved and not just second order statistics. Where CCA is ideal for Gaussian data, CIA facilitates analysis of variables with different genesis and therefore different statistical distributions. As a proof of concept we give a toy example. We also give an example with DLR 3K camera data from two time points covering a motor way.

U2 - 10.1109/Multi-Temp.2015.7245779

DO - 10.1109/Multi-Temp.2015.7245779

M3 - Article in proceedings

SN - 978-1-4673-7119-3

BT - Proceedings of the 8th International Workshop on the Analysis of Multitemporal Remote Sensing Images (Multi-Temp 2015)

PB - IEEE

T2 - 8th International Workshop on the Analysis of Multitemporal Remote Sensing Images

Y2 - 22 July 2015 through 24 July 2015

ER -

Change detection in bi-temporal data by canonical information analysis

Abstract

Workshop

Access to Document

OpenUrl availability

Fingerprint

Cite this