Detection of Object Motion Regions in Aerial Image Pairs With a Multilayer Markovian Model

We propose a new Bayesian method for detecting the regions of object displacements in aerial image pairs. We use a robust but coarse 2D image registration algorithm. Our main challenge is to eliminate the registration errors from the extracted change map. We introduce a three-layer Markov random fie...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on image processing 2009-10, Vol.18 (10), p.2303-2315
Hauptverfasser:	Benedek, C., Sziranyi, T., Kato, Z., Zerubia, J.
Format:	Artikel
Sprache:	eng
Schlagworte:	Aerial images Aerials Algorithms Applied sciences Artificial Intelligence Bayesian methods camera motion change detection Change detection algorithms Data mining Exact sciences and technology Image Enhancement - methods Image Interpretation, Computer-Assisted - methods Image processing Image registration Image segmentation Imaging, Three-Dimensional - methods Information, signal and communications theory Markov Chains Markov processes Markov random field (MRF) Markov random fields Motion Motion detection Multilayers Nonhomogeneous media Object detection Pattern Recognition, Automated - methods Photogrammetry - methods Reproducibility of Results Robustness Sensitivity and Specificity Signal processing Subtraction Technique Telecommunications and information theory Two dimensional
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	We propose a new Bayesian method for detecting the regions of object displacements in aerial image pairs. We use a robust but coarse 2D image registration algorithm. Our main challenge is to eliminate the registration errors from the extracted change map. We introduce a three-layer Markov random field ( L 3 MRF) model which integrates information from two different features, and ensures connected homogenous regions in the segmented images. Validation is given on real aerial photos.
ISSN:	1057-7149 1941-0042
DOI:	10.1109/TIP.2009.2025808