Illumination-Invariant Change Detection Using a Statistical Colinearity Criterion

Rudolf Mester, Til Aach, Lutz Dümbgen


This paper describes a new algorithm for illumination-invariant change detection that combines a simple multiplicative illumination model with decision theoretic approaches to change detection. The core of our algorithm is a new statistical test for linear dependence (colinearity) of vectors observed in noise. This criterion can be employed for a significance test, but a considerable improvement of reliability for real-world image sequences is achieved if it is integrated into a Bayesian framework that exploits spatio-temporal contiguity and prior knowledge about shape and size of typical change detection masks. In the latter approach,an MRF-based prior model for the sought change masks can be applied successfully. With this approach, spurious spot-like decision errors can be almost fully eliminated.

Original languageEnglish
Title of host publicationDAGM 2001: Pattern Recognition
Number of pages8
PublisherSpringer Verlag
Publication date01.01.2001
ISBN (Print)978-3-540-42596-0
ISBN (Electronic)978-3-540-45404-5
Publication statusPublished - 01.01.2001
Event23rd German Association for Pattern Recognition Symposium - Munich, Germany
Duration: 12.09.200114.09.2001
Conference number: 128539


Dive into the research topics of 'Illumination-Invariant Change Detection Using a Statistical Colinearity Criterion'. Together they form a unique fingerprint.

Cite this