By Topic

Anomaly detection in three dimensional data based on Gauss Markov random field modeling

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
A. Noiboar ; Dept. of Electr. Eng., Technion-Israel Inst. of Technol., Haifa, Israel ; I. Cohen

We present an anomaly detection approach for three dimensional data. We pre-process the 3D data using the Karhunen-Loeve transform (KLT), to remove correlation between data layers. Each layer is modeled as a Gauss Markov random field (GMRF). We present an efficient least squares method for model estimation. Anomaly detection is carried out in each data layer independently. We assume the anomalies lie in a known signal subspace. A different subspace is assumed for each data layer, such that a-priori knowledge about the sensors used to capture the data, or about the anomalies can be incorporated into the subspace. A parametric form of the model inverse covariance matrix is utilized to yield a computationally efficient detection. We demonstrate the performance of our approach by applying it to the detection of defects in wafer images and to detection of faults in 3D seismic data.

Published in:

Electrical and Electronics Engineers in Israel, 2004. Proceedings. 2004 23rd IEEE Convention of

Date of Conference:

6-7 Sept. 2004