Skip to main content

Object-based Change Detection Objekt-basierte Änderungsdetektion

Buy Article:

$39.00 plus tax (Refund Policy)


The iteratively reweighted multivariate alteration detection (IR-MAD) has shown to be a very useful tool for detecting changes in imagery acquired over the same area but at different times. However, applying the paradigm of object-based image analysis (OBIA) leads to the problem how to connect corresponding objects extracted from images recorded at two different times. Moreover, the huge number of object features available in OBIA results in numerical instabilities within the MAD method due to near-singular covariance matrices. The paper introduces recent developments for object-based change detection. First, a new approach to segmentation for object-based change detection will be presented: The algorithm segments the first image using the multiresolution segmentation. Assigned to the second image, all segmentation merges are checked for consistency and removed if the check fails. Second, the paper shows how to address the numerical problems in the MAD algorithm by regularisation as well as by dimensionality reduction using Principal Component Analysis (PCA). It will be demonstrated how to integrate the adapted segmentation andIR-MAD into the object-based change detection workflow.

Die Methode der Iteratively Reweighted Multivariate Alteration Detection (IR-MAD) hat sich als sehr nützliches Instrument erwiesen, um Änderungen zwischen zwei Satellitenbildern unterschiedlicher Aufnahmezeiten eines Gebietes zu analysieren. Jedoch kann die Methode nicht direkt den Ansatz der objektbasierten Bildanalyse (OBIA) integrieren. Insbesondere ist es bisher nicht möglich Änderungen der Form zwischen den Bildern zu detektieren, da die Algorithmen zur Objektextraktion, der so genannten Segmentierung, nicht robust genug sind. Darüber hinaus führt die durch das OBIA-Konzept verfügbare große Anzahl untereinander korrelierter Objekteigenschaften dazu, dass die Methode der IR-MAD numerisch instabil wird. Diese Arbeit präsentiert zwei Neuentwicklungen im Bereich der objekt-basierten Änderungsdetektion. Einerseits wird ein Algorithmus zur Bildsegmentierung vorgestellt, der es ermöglicht, die IR-MAD-Methode direkt auf Basis der Objekte durchzuführen. Andererseits zeigt diese Arbeit auf, wie die numerischen Probleme der IR-MAD-Methode durch Regularisierung und Dimensionsreduktion mittels Hauptkomponentenanalyse gelöst werden können.


Document Type: Research Article


Publication date: 2011-08-01

More about this publication?
  • Photogrammetrie - Fernerkundung - Geoinformation (PFG) is an international scholarly journal covering the progress and application of photogrammetric methods, remote sensing technology and the intricately connected field of geoinformation processing.

    Papers published in PFG highlight new developments and applications of these technologies in practice. The journal hence addresses both researchers and student of these disciplines at academic institutions and universities and the downstream users in both the private sector and public administration.

    PFG places special editorial emphasis on the communication of new methodologies in data acquisition, new approaches to optimized processing and interpretation of all types of data which were acquired by photogrammetric methods, remote sensing, image processing and the computer-aided interpretation of such data in general.

    PFG is the official journal of the German Society of Photogrammetry and Remote Sensing.
  • Editorial Board
  • Information for Authors
  • Subscribe to this Title
  • Ingenta Connect is not responsible for the content or availability of external websites
  • Access Key
  • Free content
  • Partial Free content
  • New content
  • Open access content
  • Partial Open access content
  • Subscribed content
  • Partial Subscribed content
  • Free trial content
Cookie Policy
Cookie Policy
Ingenta Connect website makes use of cookies so as to keep track of data that you have filled in. I am Happy with this Find out more